{"id":64837,"date":"2025-09-28T15:04:03","date_gmt":"2025-09-28T07:04:03","guid":{"rendered":"https:\/\/swarma.org\/?p=64837"},"modified":"2025-09-28T15:04:03","modified_gmt":"2025-09-28T07:04:03","slug":"ai%e6%a8%a1%e6%8b%9f%e7%bb%86%e8%83%9e%ef%bc%9a%e8%ae%a9%e6%b4%bb%e7%bb%86%e8%83%9e%e5%9c%a8%e5%a4%a7%e6%95%b0%e6%8d%ae%e4%b8%ad%e8%87%aa%e5%8f%91%e6%b6%8c%e7%8e%b0%ef%bc%81","status":"publish","type":"post","link":"https:\/\/swarma.org\/?p=64837","title":{"rendered":"AI\u6a21\u62df\u7ec6\u80de\uff1a\u8ba9\u6d3b\u7ec6\u80de\u5728\u5927\u6570\u636e\u4e2d\u81ea\u53d1\u6d8c\u73b0\uff01"},"content":{"rendered":"
\n
\n
\n
\n
\n
\n
\n
\n
\n
<\/section>\n<\/section>\n
\n
\n
\n
\n
\n
\n

<\/span><\/section>\n<\/section>\n
\n
\n

\u5bfc\u8bed<\/span><\/strong><\/p>\n<\/section>\n<\/section>\n

\n

<\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n

<\/strong>\u201c\u6211\u4eec\u80fd\u5728\u8ba1\u7b97\u673a\u91cc\u6784\u5efa\u4e00\u4e2a\u7ec6\u80de\u5417\uff1f\u201d \u8fd9\u4e00\u6839\u672c\u6027\u95ee\u9898\uff0c\u6b63\u968f\u7740\u4eba\u5de5\u667a\u80fd\u548c\u7ec4\u5b66\u6280\u672f\u7684\u8fdb\u6b65\u800c\u8fce\u6765\u7a81\u7834\u6027\u673a\u9047\u3002\u6700\u8fd1\uff0c\u6743\u5a01\u671f\u520aCell\u53d1\u8868\u4e86\u4e00\u7bc7\u91cd\u78c5\u7efc\u8ff0\uff0c\u6df1\u5165\u63a2\u8ba8\u4e86 \u201c\u4eba\u5de5\u667a\u80fd\u865a\u62df\u7ec6\u80de\u201d\uff08AI Virtual Cell, AIVC\uff09 \u8fd9\u4e00\u524d\u6cbf\u6982\u5ff5\u3002\u8fd9\u7bc7\u6587\u7ae0\u6c47\u96c6\u4e86AI\u4e0e\u751f\u547d\u79d1\u5b66\u9886\u57df\u7684\u9876\u5c16\u667a\u6167\u3002\u4f5c\u8005\u56e2\u961f\u9635\u5bb9\u975e\u5e38\u5f3a\u5927\uff0c\u6a2a\u8de8\u5b66\u672f\u754c\uff08\u5982\u65af\u5766\u798f\u3001\u54c8\u4f5b \u3001\u79d1\u6280\u5de8\u5934\uff08\u5982\u8c37\u6b4c\u7814\u7a76\u9662\u3001\u5fae\u8f6f\u7814\u7a76\u9662\uff09\u548c\u751f\u7269\u6280\u672f\u884c\u4e1a\uff08\u5982\u57fa\u56e0\u6cf0\u514b\uff09\uff0c\u5e76\u7531\u9648-\u624e\u514b\u4f2f\u683c\u8ba1\u5212 (CZI)\u7684\u7814\u7a76\u8005\u8054\u5408\u63a8\u52a8\uff0c\u5171\u540c\u4e3a\u6784\u5efa\u201cAI\u865a\u62df\u7ec6\u80de\u201d\u7ed8\u5236\u84dd\u56fe\u3002\u672c\u6587\u662f\u5bf9\u8be5\u7efc\u8ff0\u7684\u7ffb\u8bd1\u3002<\/span><\/span><\/strong><\/p>\n


<\/span><\/span><\/strong><\/p>\n

\u7ec6\u80de\u5bf9\u4e8e\u7406\u89e3\u5065\u5eb7\u548c\u75be\u75c5\u81f3\u5173\u91cd\u8981\uff0c\u4f46\u5176\u4f5c\u4e3a\u751f\u547d\u6700\u57fa\u672c\u529f\u80fd\u5355\u4f4d\u7684\u590d\u6742\u6027\uff0c\u8ba9\u4f20\u7edf\u5efa\u6a21\u65b9\u6cd5\u5728\u6a21\u62df\u5176\u529f\u80fd\u4e0e\u884c\u4e3a\u65f6\u663e\u5f97\u529b\u4e0d\u4ece\u5fc3\u3002\u5982\u4eca\uff0c\u6211\u4eec\u6709\u673a\u4f1a\u521b\u5efa\u4e00\u4e2a\u4eba\u5de5\u667a\u80fd\u865a\u62df\u7ec6\u80de\u2014\u2014\u4e00\u4e2a\u57fa\u4e8e\u5927\u578b\u795e\u7ecf\u7f51\u7edc\u7684\u591a\u5c3a\u5ea6\u3001\u591a\u6a21\u6001\u6a21\u578b\uff0c\u80fd\u591f\u8868\u5f81\u548c\u6a21\u62df\u5206\u5b50\u3001\u7ec6\u80de\u4e43\u81f3\u7ec4\u7ec7\u5728\u4e0d\u540c\u72b6\u6001\u4e0b\u7684\u884c\u4e3a\u3002\u8fd9\u79cdAIVC\u4e0d\u4ec5\u662f\u4e00\u79cd\u65b0\u7684\u5efa\u6a21\u5de5\u5177\uff0c\u66f4\u662f\u4e00\u79cd\u5168\u65b0\u7684\u79d1\u5b66\u8303\u5f0f\u3002<\/span><\/span><\/strong><\/p>\n


<\/span><\/span><\/strong><\/p>\n

\u901a\u8fc7\u5b9e\u73b0\u9ad8\u4fdd\u771f\u5ea6\u7684\u4eff\u771f\uff0cAIVC\u6709\u671b\u8ba9\u79d1\u5b66\u5bb6\u5728\u865a\u62df\u7a7a\u95f4\u8fdb\u884c\u590d\u6742\u7684\u201c\u8ba1\u7b97\u673a\u5b9e\u9a8c\u201d (in silico experimentation)\uff0c\u6210\u4e3a\u7406\u89e3\u75be\u75c5\u53d1\u751f\u3001\u836f\u7269\u53cd\u5e94\u548c\u7ec6\u80de\u5de5\u7a0b\u7684\u5f3a\u5927\u5de5\u5177\uff0c\u751a\u81f3\u5c06\u7ec6\u80de\u91cd\u65b0\u5b9a\u4e49\u4e3a\u4e00\u79cd\u4fe1\u606f\u5904\u7406\u7cfb\u7edf\u3002\u7136\u800c\uff0c\u6784\u5efaAIVC\u662f\u4e00\u9879\u5de8\u5927\u7684\u6311\u6218\uff0c\u9700\u8981\u6574\u5408\u8de8\u5c3a\u5ea6\u3001\u8de8\u7269\u79cd\u7684\u591a\u6837\u5316\u6570\u636e\uff0c\u5e76\u5efa\u7acb\u4e00\u4e2a\u5f00\u653e\u534f\u4f5c\u7684\u79d1\u5b66\u5171\u540c\u4f53\u3002\u5982\u540c\u5386\u53f2\u4e0a\u7684\u201c\u57fa\u56e0\u7ec4\u8ba1\u5212\u201d(human genome project)\u4e00\u6837\uff0cAIVC\u7684\u63d0\u51fa\u6807\u5fd7\u7740\u4e00\u4e2a\u91cd\u8981\u7684\u5386\u53f2\u5173\u53e3\uff0c\u5b83\u4e0d\u4ec5\u662f\u4e00\u4e2a\u5b66\u672f\u613f\u666f\uff0c\u66f4\u662f\u4e00\u79cd\u9762\u5411\u672a\u6765\u7684\u79d1\u5b66\u57fa\u7840\u8bbe\u65bd\u3002<\/span><\/span><\/strong><\/p>\n


<\/span><\/span><\/strong><\/p>\n

\u5c55\u671b\u672a\u6765\uff0c\u865a\u62df\u7ec6\u80de\u5e73\u53f0\u6216\u8bb8\u4f1a\u50cf\u663e\u5fae\u955c\u4e00\u6837\u666e\u53ca\uff0c\u5e2e\u52a9\u79d1\u5b66\u5bb6\u9ad8\u6548\u5730\u9a8c\u8bc1\u5047\u8bbe\u3001\u8bbe\u8ba1\u5b9e\u9a8c\u3002\u8fd9\u7bc7\u7efc\u8ff0\u6240\u63cf\u7ed8\u7684\uff0c\u6b63\u662f\u4e00\u5e45\u5173\u4e8e\u672a\u6765\u751f\u7269\u5b66\u7814\u7a76\u5982\u4f55\u88ab\u6df1\u523b\u91cd\u5851\u7684\u5ba3\u8a00\u3002\u5176\u6700\u7ec8\u76ee\u6807\u5e76\u975e\u66ff\u4ee3\u81ea\u7136\uff0c\u800c\u662f\u4e3a\u4e86\u4e0e\u751f\u547d\u4e16\u754c\u5c55\u5f00\u66f4\u79d1\u5b66\u3001\u9ad8\u6548\u7684\u5bf9\u8bdd\u3002<\/span><\/span><\/strong><\/p>\n


<\/span><\/span><\/strong><\/p>\n

\u5173\u952e\u8bcd\uff1a\u7ec6\u80de\u751f\u7269\u5b66\u3001\u4eba\u5de5\u667a\u80fd\u3001\u673a\u5668\u5b66\u4e60\u3001\u865a\u62df\u7ec6\u80de<\/span><\/strong><\/strong><\/span><\/strong><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

\n
\n
\n
<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\"\u56fe\u7247\"<\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

\u738b\u660a\u4e28\u4f5c\u8005<\/span><\/span>
<\/span><\/p>\n

\u5468\u8389<\/span>\uff5c\u5ba1\u6821<\/span><\/span><\/p>\n


<\/span><\/p>\n

<\/section>\n
\n

\u8bba\u6587\u9898\u76ee\uff1aHow to build the virtual cell with artificial intelligence: Priorities and opportunities<\/span><\/p>\n

\u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/www.cell.com\/cell\/fulltext\/S0092-8674(24)01332-1<\/span><\/p>\n

\u8bba\u6587\u6765\u6e90\uff1aCell<\/span><\/p>\n<\/blockquote>\n


<\/span><\/p>\n

\n
\n
\n
\n
\n
\n
\u76ee\u5f55<\/span><\/p>\n
\n
\n
\n
    \n
  1. \n

    \u6458\u8981<\/span><\/span><\/p>\n<\/li>\n

  2. \n

    \u5f15\u8a00<\/span><\/span><\/p>\n<\/li>\n

  3. \n

    \u5efa\u7acbAIVC\u6240\u9762\u4e34\u7684\u91cd\u5927\u6311\u6218<\/span><\/span><\/p>\n<\/li>\n

  4. \n

    AIVCs\u7684\u5173\u952e\u80fd\u529b<\/span><\/span><\/p>\n<\/li>\n

  5. \n

    \u6784\u5efaAIVC\u7684\u65b9\u6cd5<\/span><\/span><\/p>\n<\/li>\n

  6. \n

    \u6784\u5efaAIVC\u7684\u6570\u636e\u9700\u6c42\u548c\u8981\u6c42<\/span><\/span><\/p>\n<\/li>\n

  7. \n

    AIVC\u7684\u6a21\u578b\u8bc4\u4f30\u548c\u53ef\u89e3\u91ca\u6027<\/span><\/span><\/p>\n<\/li>\n

  8. \n

    \u524d\u666f\u5c55\u671b<\/span><\/span><\/p>\n<\/li>\n<\/ol>\n<\/section>\n<\/section>\n<\/section>\n

    <\/strong><\/section>\n

    <\/strong><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n


    <\/span><\/p>\n

    \u7ec6\u80de\u5bf9\u4e8e\u7406\u89e3\u5065\u5eb7\u548c\u75be\u75c5\u81f3\u5173\u91cd\u8981\uff0c\u7136\u800c\u4f20\u7edf\u6a21\u578b\u5728\u5efa\u6a21\u548c\u6a21\u62df\u7ec6\u80de\u529f\u80fd\u4e0e\u884c\u4e3a\u65b9\u9762\u5b58\u5728\u4e0d\u8db3\u3002\u4eba\u5de5\u667a\u80fd\u548c\u7ec4\u5b66\u6280\u672f\u7684\u8fdb\u6b65\u4e3a\u521b\u5efa AI \u865a\u62df\u7ec6\u80de<\/span>\uff08AIVC\uff09<\/span>\u63d0\u4f9b\u4e86\u7a81\u7834\u6027\u673a\u9047\uff0c\u8fd9\u662f\u4e00\u4e2a\u57fa\u4e8e\u5927\u578b\u795e\u7ecf\u7f51\u7edc\u7684\u591a\u5c3a\u5ea6\u3001\u591a\u6a21\u6001\u6a21\u578b<\/span> (a multi-scale, multi-modal large-neural-network-based model)<\/span>\uff0c\u80fd\u591f\u8868\u5f81\u548c\u6a21\u62df\u5206\u5b50\u3001\u7ec6\u80de\u548c\u7ec4\u7ec7\u5728\u4e0d\u540c\u72b6\u6001\u4e0b\u7684\u884c\u4e3a\u3002\u672c\u6587\u63d0\u51fa\u4e86 AIVC \u8bbe\u8ba1\u7684\u613f\u666f\uff0c\u4ee5\u53ca\u6784\u5efa AIVC \u7684\u534f\u4f5c\u52aa\u529b\u5c06\u5982\u4f55\u901a\u8fc7\u5b9e\u73b0\u9ad8\u4fdd\u771f\u5ea6\u4eff\u771f\u3001\u52a0\u901f\u79d1\u5b66\u53d1\u73b0\u548c\u6307\u5bfc\u5b9e\u9a8c\u7814\u7a76\u6765\u53d8\u9769\u751f\u7269\u5b66\u7814\u7a76\uff0c\u4e3a\u7406\u89e3\u7ec6\u80de\u529f\u80fd\u548c\u4fc3\u8fdb\u5f00\u653e\u79d1\u5b66\u4e2d\u7684\u8de8\u5b66\u79d1\u5408\u4f5c\u63d0\u4f9b\u65b0\u673a\u9047\u3002<\/span><\/span><\/p>\n


    <\/span><\/p>\n

    \n
    \n
    \n
    \n
    \n

    <\/span><\/section>\n

    <\/span><\/section>\n
    \n
    <\/section>\n<\/section>\n<\/section>\n
    \n

    1. \u80cc\u666f<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

    \n

    <\/span><\/section>\n

    <\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


    <\/span><\/p>\n

    \u7ec6\u80de\uff0c\u4f5c\u4e3a\u751f\u547d\u7684\u57fa\u672c\u5355\u4f4d\uff0c\u662f\u4e00\u4e2a\u6781\u5176\u590d\u6742\u7684\u5b9e\u4f53\uff0c\u5176\u6027\u8d28\u548c\u884c\u4e3a\u6311\u6218\u7740\u7269\u7406\u548c\u8ba1\u7b97\u5efa\u6a21\u7684\u6781\u9650\u3002\u6bcf\u4e2a\u7ec6\u80de\u90fd\u662f\u4e00\u4e2a\u52a8\u6001\u7684\u9002\u5e94\u6027\u7cfb\u7edf\uff0c\u5176\u590d\u6742\u7684\u884c\u4e3a\u4ece\u65e0\u6570\u5206\u5b50\u76f8\u4e92\u4f5c\u7528\u4e2d\u6d8c\u73b0\u51fa\u6765\u3002\u67d0\u4e9b\u65b9\u9762\u5bf9\u6270\u52a8\u5177\u6709\u663e\u8457\u7684\u9c81\u68d2\u6027\uff0c\u6bd4\u5982\u57fa\u56e0\u7684\u7f3a\u5931\u6216\u7528\u6765\u81ea\u4e0d\u540c\u7269\u79cd\u7684\u540c\u6e90\u57fa\u56e0\u66ff\u6362\u3002\u800c\u5176\u4ed6\u65b9\u9762\u5219\u5bf9\u770b\u4f3c\u5fae\u5c0f\u7684\u5e72\u6270\u90fd\u5f88\u654f\u611f\uff0c\u6bd4\u5982\u70b9\u7a81\u53d8\u6216\u5916\u90e8\u56e0\u5b50\uff0c\u8fd9\u4e9b\u90fd\u53ef\u80fd\u4f7f\u7ec6\u80de\u9677\u5165\u529f\u80fd\u5931\u8c03\u548c\u75be\u75c5\u72b6\u6001\u3002<\/span><\/span><\/p>\n

    \u4e3a\u4e86\u7406\u89e3\u7ec6\u80de\u7684\u529f\u80fd\uff0c\u79d1\u5b66\u5bb6\u4eec\u8bd5\u56fe\u6784\u5efa\u865a\u62df\u7ec6\u80de\u6a21\u578b<\/span>\uff08virtual cell models\uff09<\/span>\uff0c\u4ee5\u6a21\u62df\u3001\u9884\u6d4b\u548c\u8c03\u63a7\u7ec6\u80de\u884c\u4e3a [1\u20136]\u3002\u73b0\u6709\u7684\u7ec6\u80de\u6a21\u578b\u901a\u5e38\u57fa\u4e8e\u89c4\u5219<\/span>\uff08rule-based\uff09<\/span>\uff0c\u5c06\u5bf9\u6f5c\u5728\u751f\u7269\u673a\u5236\u7684\u5047\u8bbe\u4e0e\u4ece\u89c2\u6d4b\u6570\u636e\u4e2d\u62df\u5408\u7684\u53c2\u6570\u76f8\u7ed3\u5408\u3002\u8fd9\u4e9b\u6a21\u578b\u5f80\u5f80\u4f9d\u8d56\u4e8e\u660e\u786e\u5b9a\u4e49\u7684\u6570\u5b66\u6216\u8ba1\u7b97\u65b9\u6cd5\uff0c\u4f8b\u5982\u5fae\u5206\u65b9\u7a0b [7\u20139]\u3001\u968f\u673a\u6a21\u62df [10,11] \u6216\u57fa\u4e8e\u667a\u80fd\u4f53\u7684\u6a21\u578b [12,13]\u3002\u5b83\u4eec\u5728\u590d\u6742\u6027\u4e0a\u5b58\u5728\u5dee\u5f02\uff0c\u6db5\u76d6\u7ec6\u80de\u751f\u7269\u5b66\u4e2d\u591a\u4e2a\u660e\u786e\u7684\u7814\u7a76\u65b9\u5411\uff0c\u5982\u8f6c\u5f55\u4e0e\u7ffb\u8bd1 [14]\u3001\u7ec6\u80de\u9aa8\u67b6\u9a71\u52a8\u7684\u7ec6\u80de\u884c\u4e3a [15,16]\u3001\u751f\u5316\u7f51\u7edc [17] \u4ee5\u53ca\u4ee3\u8c22\u901a\u91cf [18,19]\u3002\u9996\u4e2a\u5168\u7ec6\u80de\u6a21\u578b\u4e8e 2012 \u5e74\u5f00\u53d1\uff0c\u5176\u4ee3\u8868\u4e86\u751f\u6b96\u652f\u539f\u4f53<\/span>\uff08Mycobacterium genitalium\uff09<\/span>\u8fd9\u4e00\u751f\u7269\u4f53\u4e2d\u5df2\u77e5\u7684\u5168\u90e8 482 \u4e2a\u57fa\u56e0\u53ca\u5206\u5b50\u529f\u80fd [8]\u3002\u81ea\u8fd9\u9879\u5f00\u521b\u6027\u5de5\u4f5c\u4ee5\u6765\uff0c\u7814\u7a76\u8005\u4eec\u5df2\u5f00\u53d1\u51fa\u5168\u57fa\u56e0\u7ec4\u6a21\u578b\uff0c\u7528\u4ee5\u4ee3\u8868\u5305\u62ec\u5927\u80a0\u6746\u83cc\u5728\u5185\u7684\u5176\u4ed6\u7ec6\u83cc\u751f\u7269\u4f53 [8,20\u201322]\u3002<\/span><\/span><\/p>\n

    \u5c3d\u7ba1\u8fd9\u4e9b\u65b9\u6cd5\u5728\u751f\u7269\u7cfb\u7edf\u5efa\u6a21\u4e2d\u5e7f\u6cdb\u5e94\u7528\uff0c\u4f46\u8fc4\u4eca\u4e3a\u6b62\u7684\u65b9\u6cd5\u90fd\u65e0\u6cd5\u5145\u5206\u6355\u6349\u7ec6\u83cc\u548c\u66f4\u590d\u6742\u7cfb\u7edf<\/span>\uff08\u5982\u4eba\u7c7b\u7ec6\u80de\uff09<\/span>\u8fd0\u884c\u7684\u8bb8\u591a\u65b9\u9762\u3002 \u8fd9\u4e9b\u6311\u6218\u5171\u540c\u6784\u6210\u4e86\u6784\u5efa\u865a\u62df\u7ec6\u80de\u7684\u91cd\u5927\u969c\u788d\uff0c\u4f53\u73b0\u5728\u591a\u4e2a\u5c42\u9762\u3002 \u9996\u5148\uff0c\u5728<\/span><\/span>\u591a\u5c3a\u5ea6\u5efa\u6a21 (Multi-scale modeling\uff09<\/span><\/span><\/strong>\u65b9\u9762\uff0c\u7ec6\u80de\u5728\u65f6\u95f4\u548c\u7a7a\u95f4\u4e0a\u8de8\u8d8a\u4ece\u539f\u5b50\u3001\u5206\u5b50\u5230\u7ec6\u80de\u548c\u7ec4\u7ec7\u7684\u591a\u4e2a\u5c3a\u5ea6\u8fd0\u884c\uff0c\u5176\u529f\u80fd\u7279\u6027\u901a\u8fc7\u5c3a\u5ea6\u95f4\u7684\u975e\u7ebf\u6027\u8f6c\u6362\u800c\u4ea7\u751f \u3002\u5176\u6b21\uff0c\u7ec6\u80de\u529f\u80fd\u5305\u542b<\/span><\/span>\u4f17\u591a\u76f8\u4e92\u4f5c\u7528\u4e14\u591a\u6837\u5316\u7684\u8fc7\u7a0b (Diverse processes with massive numbers of interacting components)<\/span><\/span><\/strong>\uff0c\u5982\u57fa\u56e0\u8c03\u63a7\u3001\u4ee3\u8c22\u548c\u4fe1\u53f7\u8f6c\u5bfc<\/span>\uff08transduction\uff09<\/span>\uff0c\u6bcf\u4e2a\u8fc7\u7a0b\u90fd\u6d89\u53ca\u5927\u91cf\u5904\u4e8e\u52a8\u6001\u6784\u578b\u4e2d\u7684\u751f\u7269\u5206\u5b50\u79cd\u7c7b \u3002\u6700\u540e\uff0c\u5927\u591a\u6570\u7ec6\u80de\u8fc7\u7a0b\u8868\u73b0\u51fa<\/span><\/span>\u9ad8\u5ea6\u7684\u975e\u7ebf\u6027\u52a8\u529b\u5b66 (Nonlinear dynamics)<\/span><\/span><\/strong>\uff0c\u8f93\u5165\u7684\u5fae\u5c0f\u53d8\u5316\u53ef\u80fd\u5bfc\u81f4\u8f93\u51fa\u7684\u590d\u6742\u53d8\u5316\uff0c\u8fd9\u8fdb\u4e00\u6b65\u589e\u52a0\u4e86\u5efa\u6a21\u7684\u96be\u5ea6 \u3002<\/span><\/span><\/span><\/p>\n

    \u79d1\u5b66\u6280\u672f\u4e2d\u7684\u4e24\u6b21\u6fc0\u52a8\u4eba\u5fc3\u7684\u9769\u547d\u2014\u2014\u4eba\u5de5\u667a\u80fd\u9769\u547d\u548c\u7ec4\u5b66\u9769\u547d\u2014\u2014\u4f7f\u5f97\u76f4\u63a5\u4ece\u6570\u636e\u4e2d\u5b66\u4e60\u6784\u5efa\u7ec6\u80de\u6a21\u578b\u6210\u4e3a\u53ef\u80fd\u3002\u8fd9\u4e24\u6b21\u5e76\u884c\u7684\u9769\u547d\u4e3a\u4e00\u4e2a\u96c4\u5fc3\u52c3\u52c3\u7684\u613f\u666f\u63d0\u4f9b\u4e86\u524d\u6240\u672a\u6709\u7684\u673a\u9047\uff0c\u5373\u6784\u5efa <\/span><\/span>AI \u865a\u62df\u7ec6\u80de (AIVC)<\/span><\/span><\/strong><\/em>\uff0c\u8fd9\u662f\u4e00\u4e2a\u57fa\u4e8e\u5927\u578b\u795e\u7ecf\u7f51\u7edc\u7684\u591a\u5c3a\u5ea6\u3001\u591a\u6a21\u6001\u6a21\u578b\uff0c\u80fd\u591f\u8868\u5f81\u548c\u6a21\u62df\u5206\u5b50\u3001\u7ec6\u80de\u548c\u7ec4\u7ec7\u5728\u4e0d\u540c\u72b6\u6001\u4e0b\u7684\u884c\u4e3a<\/span>\uff08\u56fe 1\uff09<\/span>\u3002<\/span><\/span><\/span><\/p>\n

    \u5728\u5b9e\u9a8c\u65b9\u9762\uff0c\u6d4b\u91cf\u6280\u672f\u901a\u91cf\u7684\u6307\u6570\u7ea7\u589e\u957f\u4f7f\u5f97\u5728\u4e0d\u540c\u7ec6\u80de\u548c\u7ec4\u7ec7\u7cfb\u7edf\u5185\u90e8\u53ca\u4e4b\u95f4\u6536\u96c6\u4e86\u5927\u91cf\u4e14\u4e0d\u65ad\u589e\u957f\u7684\u53c2\u8003\u6570\u636e\u96c6 [23\u201325]\uff0c\u5728\u8fc7\u53bb\u51e0\u5e74\u4e2d\u6570\u636e\u6bcf 6 \u4e2a\u6708\u7ffb\u4e00\u756a [26]\uff0c\u540c\u65f6\u8fd9\u4e9b\u6d4b\u91cf\u8fd8\u80fd\u4e0e\u7cfb\u7edf\u6027\u6270\u52a8\u76f8\u7ed3\u5408[27\u201329]\u3002\u5728\u8ba1\u7b97\u65b9\u9762\uff0c\u4eba\u5de5\u667a\u80fd\u6280\u672f\u7684\u5feb\u901f\u53d1\u5c55\u589e\u5f3a\u4e86\u6211\u4eec\u76f4\u63a5\u4ece\u6570\u636e\u4e2d\u5b66\u4e60\u6a21\u5f0f\u548c\u8fc7\u7a0b\u7684\u80fd\u529b\uff0c\u800c\u65e0\u9700\u660e\u786e\u7684\u89c4\u5219\u6216\u4eba\u5de5\u6807\u6ce8 [30,31]\u3002\u8fd9\u79cd\u5efa\u6a21\u8303\u5f0f\u5df2\u5728\u751f\u7269\u5206\u5b50\u9886\u57df\u5f97\u5230\u6210\u529f\u5e94\u7528\uff0c\u4f8b\u5982\uff0c\u4ece\u5e8f\u5217\u9884\u6d4b\u4e09\u7ef4<\/span> (3D) <\/span>\u5206\u5b50\u7ed3\u6784[32\u201334] \u4ee5\u53ca\u4e0d\u540c\u5206\u5b50\u7ec4\u5206\u4e4b\u95f4\u7684\u76f8\u4e92\u4f5c\u7528 [35\u201338]\u3002\u5176\u4e2d\uff0cGoogle DeepMind \u7684 AlphaFold \u5728\u8be5\u9886\u57df\u53d6\u5f97\u4e86\u7a81\u7834\u6027\u8fdb\u5c55\u3002\u4eba\u5de5\u667a\u80fd\u9886\u57df\u6700\u65b0\u7684\u5efa\u6a21\u65b9\u6cd5\u63d0\u4f9b\u4e86\u8868\u5f81\u548c\u63a8\u7406\u5de5\u5177\uff0c\u6ee1\u8db3\u4e86\u9884\u6d4b\u6027<\/span> (predictive)<\/span>\u3001\u751f\u6210\u6027<\/span> (generative) <\/span>\u548c\u53ef\u67e5\u8be2\u6027<\/span> (queryable\uff09<\/span>\u8fd9\u4e09\u4e2a\u5173\u952e\u7279\u6027\uff0c\u8fd9\u4e9b\u90fd\u662f\u63a8\u8fdb\u751f\u7269\u5b66\u7814\u7a76\u548c\u7406\u89e3\u7684\u5173\u952e\u529f\u80fd\u3002<\/span><\/span><\/p>\n

    \u57fa\u4e8e\u8fd9\u4e9b\u7279\u6027\uff0c\u6211\u4eec\u8ba4\u4e3a\u73b0\u5728\u5df2\u7ecf\u5177\u5907\u4e86\u5f00\u53d1\u5b8c\u5168\u6570\u636e\u9a71\u52a8\u7684\u57fa\u4e8e\u795e\u7ecf\u7f51\u7edc\u7684 AIVC \u8868\u5f81\u7684\u65b9\u6cd5\uff0c\u8fd9\u53ef\u4ee5\u901a\u8fc7\u5b9e\u73b0\u5feb\u8282\u594f\u7684\u8ba1\u7b97\u673a\u6a21\u62df\u7814\u7a76\u4ee5\u53ca\u5728\u8ba1\u7b97\u65b9\u6cd5\u548c\u9a8c\u8bc1\u6027\u6e7f\u5b9e\u9a8c\u5ba4\u5b9e\u9a8c\u4e4b\u95f4\u5efa\u7acb\u5f3a\u5927\u6865\u6881\u6765\u52a0\u901f\u751f\u7269\u533b\u5b66\u7814\u7a76<\/span>\uff08\u56fe 1\uff09<\/span>\u3002<\/span><\/span><\/p>\n

    <\/section>\n

    \u56fe 1. AIVC \u7684\u80fd\u529b<\/span><\/span><\/p>\n

    (A) AIVC \u63d0\u4f9b\u4e86\u4e00\u79cd\u901a\u7528\u8868\u793a (universal representation, UR)\uff0c\u53ef\u4ee5\u8de8\u7269\u79cd\u548c\u6761\u4ef6\u83b7\u5f97\u7ec6\u80de\u72b6\u6001\uff0c\u5e76\u53ef\u4ece\u8de8\u5c3a\u5ea6 (\u5206\u5b50\u3001\u7ec6\u80de\u548c\u591a\u7ec6\u80de) \u7684\u4e0d\u540c\u6570\u636e\u6a21\u6001\u4e2d\u751f\u6210\u3002<\/span><\/span><\/span><\/p>\n

    (B) AIVC \u5177\u5907\u8868\u793a\u548c\u9884\u6d4b\u7ec6\u80de\u751f\u7269\u5b66\u7279\u6027\u7684\u80fd\u529b\u3002\u8fd9\u79cd\u901a\u7528\u6027\u4f7f\u8be5\u8868\u793a\u80fd\u591f\u4f5c\u4e3a\u53c2\u8003\u57fa\u51c6\uff0c\u6cdb\u5316\u5230\u6b64\u524d\u672a\u89c2\u5bdf\u5230\u7684\u7ec6\u80de\u72b6\u6001\uff0c\u4e3a\u672a\u6765\u7684\u6570\u636e\u751f\u6210\u63d0\u4f9b\u6307\u5bfc\u3002\u7531\u4e8e\u8be5\u8868\u793a\u5728\u4e0d\u540c\u6a21\u6001\u95f4\u5171\u4eab\uff0c\u5176\u751f\u6210\u4e0d\u53d7\u7279\u5b9a\u6570\u636e\u7c7b\u578b\u7684\u9650\u5236\uff0c\u53ef\u4f5c\u4e3a\u8de8\u6a21\u6001\u7edf\u4e00\u5206\u6790\u7684\u865a\u62df\u8868\u793a\u3002AIVC \u8fd8\u80fd\u591f\u5bf9\u7ec6\u80de\u5728\u4e0d\u540c\u72b6\u6001\u95f4\u8f6c\u6362\u7684\u52a8\u529b\u5b66\u8fc7\u7a0b\u8fdb\u884c\u5efa\u6a21\uff0c\u65e0\u8bba\u662f\u5206\u5316\u7b49\u81ea\u7136\u8fc7\u7a0b\u3001\u9057\u4f20\u53d8\u5f02\u5bfc\u81f4\u7684\u8f6c\u6362\uff0c\u8fd8\u662f\u901a\u8fc7\u5de5\u7a0b\u6270\u52a8\u4eba\u4e3a\u5f15\u53d1\u7684\u8f6c\u6362\u3002\u56e0\u6b64\uff0cAIVC \u80fd\u591f\u5b9e\u73b0\u8ba1\u7b97\u673a\u6a21\u62df\u5b9e\u9a8c\uff0c\u800c\u8fd9\u7c7b\u5b9e\u9a8c\u5728\u5b9e\u9a8c\u5ba4\u4e2d\u53ef\u80fd\u6210\u672c\u8fc7\u9ad8\u6216\u6839\u672c\u65e0\u6cd5\u5f00\u5c55\u3002<\/span><\/span><\/p>\n

    \uff08C\uff09AI\u865a\u62df\u7ec6\u80de\uff08AIVC\uff09\u7684\u6700\u7ec8\u4ef7\u503c\uff0c\u53d6\u51b3\u4e8e\u5176\u5728\u4e0d\u540c\u5c42\u9762\u4e0e\u4eba\u7c7b\u7684\u6709\u6548\u4e92\u52a8\u3002\u5bf9\u4e8e<\/span><\/span>\u4e2a\u4f53\u79d1\u5b66\u5bb6<\/span><\/span><\/strong>\u800c\u8a00\uff0c\u5176\u6f5c\u529b\u5fc5\u987b\u901a\u8fc7\u5f00\u653e\u7684\u8bb8\u53ef\u8bc1\u548c\u666e\u60e0\u7684\u8ba1\u7b97\u8d44\u6e90\u6765\u5145\u5206\u91ca\u653e\uff0c\u786e\u4fdd\u5e7f\u6cdb\u7684\u53ef\u53ca\u6027 \u3002\u540c\u65f6\uff0c\u53ef\u4ee5\u501f\u52a9\u8bed\u8a00\u6a21\u578b\u7b49\u4e2d\u4ecb\u5c42\u6765\u589e\u5f3a\u5176\u53ef\u89e3\u91ca\u6027\uff0c\u8ba9\u865a\u62df\u7ec6\u80de\u80fd\u6e05\u6670\u5730\u4f20\u8fbe\u5176\u590d\u6742\u7684\u5206\u6790\u7ed3\u679c\u3002\u5728\u66f4\u5e7f\u6cdb\u7684<\/span><\/span>\u79d1\u5b66\u754c<\/span><\/span><\/strong>\u5c42\u9762\uff0c\u5bf9AIVC\u7684\u8bc4\u4f30\u4e0d\u5e94\u5c40\u9650\u4e8e\u72ed\u9698\u7684\u57fa\u51c6\u6d4b\u8bd5\uff0c\u800c\u5e94\u805a\u7126\u4e8e\u5176\u6838\u5fc3\u80fd\u529b\u7684\u9a8c\u8bc1 \u3002\u4e00\u4e2a\u5f00\u653e\u3001\u6d3b\u8dc3\u7684\u793e\u533a\u751f\u6001\u5bf9\u4e8e\u6a21\u578b\u7684\u6301\u7eed\u8fed\u4ee3\u3001\u6539\u8fdb\u548c\u4fdd\u6301\u5f00\u653e\u6027\u81f3\u5173\u91cd\u8981 \u3002\u6700\u7ec8\uff0c\u5728<\/span><\/span>\u793e\u4f1a\u5c42\u9762<\/span><\/span><\/strong><\/span>\uff0cAIVC\u5fc5\u987b\u5c06\u6570\u636e\u9690\u79c1\u548c\u5b89\u5168\u4f5c\u4e3a\u57fa\u77f3\uff0c\u901a\u8fc7\u5f3a\u5927\u7684\u9690\u79c1\u4fdd\u62a4\u63aa\u65bd\u6765\u5b88\u62a4\u654f\u611f\u4fe1\u606f\uff0c\u4ece\u800c\u5efa\u7acb\u516c\u4f17\u4fe1\u4efb\u3002<\/span><\/span><\/span><\/p>\n


    <\/span><\/h3>\n

    \n
    \n
    \n
    \n
    \n

    <\/span><\/section>\n

    <\/span><\/section>\n
    \n
    <\/section>\n<\/section>\n<\/section>\n
    \n

    2. \u5efa\u7acbAIVC\u6240\u9762\u4e34\u7684\u91cd\u5927\u6311\u6218<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

    \n

    <\/span><\/section>\n

    <\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


    <\/span><\/p>\n

      \n
    • \n

      \u6982\u8ff0\u80fd\u529b\u5e76\u8bbe\u8ba1\u8bc4\u4f30\u6846\u67b6<\/span><\/span><\/strong><\/p>\n<\/li>\n<\/ul>\n

      \u751f\u7269\u5b66\u9886\u57df\u4e2d\u6570\u91cf\u6fc0\u589e\u7684\u57fa\u7840\u6a21\u578b<\/span> (foundation models) <\/span>\u6267\u884c\u7740\u672c\u6587\u6240\u6982\u8ff0\u7684\u865a\u62df\u7ec6\u80de\u80fd\u529b\u7684\u4e00\u4e2a\u5b50\u96c6\u3002\u9274\u4e8e\u8fd9\u4e9b\u65b9\u6cd5\u7684\u591a\u6837\u6027\uff0c\u5b9a\u4e49 AIVC \u7684\u6838\u5fc3\u80fd\u529b\u5e94\u8be5\u662f\u4ec0\u4e48\u4ee5\u53ca\u5982\u4f55\u8bc4\u4f30\u8fd9\u4e9b\u80fd\u529b\u53d8\u5f97\u975e\u5e38\u91cd\u8981\u3002\u5bf9\u4e8e\u6bcf\u4e00\u79cd\u80fd\u529b\uff0c\u90fd\u5fc5\u987b\u8bbe\u8ba1\u9002\u5f53\u7684\u6307\u6807\uff0c\u5e76\u6536\u96c6\u5168\u9762\u7684\u8bc4\u4f30\u6570\u636e\u3002\u6a21\u578b\u7684\u80fd\u529b\u5e94\u8be5\u5728\u4e00\u822c\u6027\u80fd\u4ee5\u53ca\u56de\u7b54\u7279\u5b9a\u751f\u7269\u5b66\u95ee\u9898\u7684\u80fd\u529b\u65b9\u9762\u8fdb\u884c\u8bc4\u4f30\u3002\u6301\u7eed\u6539\u8fdb\u57fa\u51c6\u6d4b\u8bd5\u7b56\u7565\u4e0e AIVC \u6a21\u578b\u540c\u6b65\u53d1\u5c55\uff0c\u5e76\u786e\u4fdd\u5b83\u4eec\u4e0e\u751f\u7269\u5b66\u4e0a\u6709\u610f\u4e49\u7684\u76ee\u6807\u4fdd\u6301\u4e00\u81f4\uff0c\u8fd9\u4e00\u70b9\u81f3\u5173\u91cd\u8981\u3002\u968f\u7740\u8be5\u9886\u57df\u5728\u8fd9\u4e9b\u95ee\u9898\u4e0a\u8fbe\u6210\u66f4\u597d\u7684\u5171\u8bc6\uff0c\u5408\u4f5c\u673a\u4f1a\u5c06\u4f1a\u51fa\u73b0\uff0c\u751f\u6210\u865a\u62df\u7ec6\u80de\u7684\u901f\u5ea6\u4e5f\u5c06\u52a0\u5feb\u3002<\/span><\/span><\/p>\n

        \n
      • \n

        \u5728\u4e0d\u540c\u67b6\u6784\u7684\u53d8\u5316\u60c5\u5883\u4e2d\u5efa\u7acb\u81ea\u4e00\u81f4\u6027<\/span><\/span><\/strong><\/p>\n<\/li>\n<\/ul>\n

        \u751f\u7269\u5b66\u6781\u5176\u590d\u6742\uff1a\u5b83\u5728\u4e0d\u540c\u5c3a\u5ea6\u3001\u4e0d\u540c\u73af\u5883\u4e2d\u8fd0\u4f5c\uff0c\u5e76\u901a\u8fc7\u4e0d\u540c\u6a21\u5f0f\u8fdb\u884c\u6d4b\u91cf\u3002AIVC \u6a21\u578b\u5fc5\u987b\u5728\u6240\u6709\u8fd9\u4e9b\u7ef4\u5ea6\u4e0a\u4fdd\u6301\u81ea\u6d3d\u6027<\/span> (self-consistent) <\/span>\u3002  \u5177\u4f53\u6765\u8bf4\uff0c\u8fd9\u79cd\u81ea\u6d3d\u6027\u4f53\u73b0\u5728\u4e24\u4e2a\u5c42\u9762\uff1a\u8de8\u7269\u7406\u5c3a\u5ea6\u7684\u4e00\u81f4\u6027\uff1a\u6a21\u578b\u7684\u529f\u80fd\u5fc5\u987b\u80fd\u5728\u4e0d\u540c\u7269\u7406\u5c3a\u5ea6\u95f4\u8d2f\u7a7f\u4f53\u73b0\u3002\u4f8b\u5982\uff0c\u5206\u5b50\u95f4\u7684\u76f8\u4e92\u4f5c\u7528\uff0c\u65e0\u8bba\u662f\u5728\u6d4b\u91cf\u5176<\/span><\/span>\u7ed3\u5408\u4eb2\u548c\u529b (binding affinity)<\/span><\/span><\/strong>\uff0c\u8fd8\u662f\u5728\u89c2\u5bdf\u5176\u5bf9\u57fa\u56e0\u8868\u8fbe\u3001\u7ec6\u80de\u95f4\u901a\u8baf\u6216\u7ec4\u7ec7\u7ed3\u6784\u7684\u5f71\u54cd\u65f6\uff0c\u90fd\u5e94\u4ea7\u751f\u903b\u8f91\u4e00\u81f4\u4e14\u76f8\u4e92\u543b\u5408\u7684\u6548\u5e94\u3002\u4ee5\u53ca\u8de8\u6570\u636e\u6a21\u5f0f\u7684\u4e00\u81f4\u6027\uff1a\u6a21\u578b\u7684\u9884\u6d4b\u7ed3\u679c\u5e94\u5f53\u72ec\u7acb\u4e8e\u5176\u8f93\u5165\u548c\u8f93\u51fa\u7684\u6570\u636e\u6a21\u5f0f\u3002\u8fd9\u610f\u5473\u7740\uff0c\u540c\u4e00\u4e2a\u751f\u7269\u5b9e\u4f53\uff0c\u5373\u4f7f\u7528\u4e0d\u540c\u7684\u6280\u672f\u8fdb\u884c\u5206\u6790\uff0c\u5728AIVC\u4e2d\u4e5f\u5e94\u88ab\u8d4b\u4e88\u7edf\u4e00\u7684\u5185\u90e8\u8868\u5f81\u3002<\/span><\/span><\/span><\/p>\n

          \n
        • \n

          \u5e73\u8861\u53ef\u89e3\u91ca\u6027\u548c\u751f\u7269\u5b66\u6548\u7528<\/span><\/span><\/strong><\/p>\n<\/li>\n<\/ul>\n

          \u6df1\u5ea6\u5b66\u4e60\u65b9\u6cd5\u5728\u751f\u7269\u5b66\u4e2d\u5e94\u7528\u7684\u4e00\u4e2a\u4e00\u81f4\u8d8b\u52bf\u662f\uff0c\u5728\u5927\u578b\u57fa\u7840\u6a21\u578b\u5174\u8d77\u7684\u63a8\u52a8\u4e0b\uff0c\u6a21\u578b\u6027\u80fd\u63d0\u5347\u4e0e\u5176\u65e5\u76ca\u4e0d\u53ef\u89e3\u91ca\u7684 \u201c\u9ed1\u76d2\u201d \u7279\u6027\u4e4b\u95f4\u5b58\u5728\u9690\u6027\u6743\u8861\u3002AIVC \u6a21\u578b\u6700\u7ec8\u5c06\u6839\u636e\u5176\u6269\u5c55\u6211\u4eec\u5bf9\u751f\u7269\u5b66\u7406\u89e3\u7684\u80fd\u529b\u6765\u8bc4\u5224\uff0c\u65e0\u8bba\u662f\u901a\u8fc7\u4e3a\u751f\u7269\u8fc7\u7a0b\u63d0\u4f9b\u65b0\u9896\u89c1\u89e3\uff0c\u8fd8\u662f\u901a\u8fc7\u52a0\u901f\u79d1\u5b66\u7814\u7a76\u8fc7\u7a0b\u3002<\/span><\/span><\/p>\n

          \u4e3a\u5b9e\u73b0\u8fd9\u4e00\u76ee\u6807\uff0cAIVC \u6a21\u578b\u5fc5\u987b\u505a\u51fa\u9ad8\u5ea6\u51c6\u786e\u4e14\u6821\u51c6\u826f\u597d\u7684\u9884\u6d4b\u6765\u6a21\u62df\u751f\u7269\u5b66\uff0c\u5e76\u4e14\u5fc5\u987b\u5e73\u8861\u53ef\u64cd\u4f5c\u6027\u4e0e\u53ef\u89e3\u91ca\u6027\u4e4b\u95f4\u7684\u6743\u8861\u3002\u53ef\u64cd\u4f5c\u7684\u6a21\u578b\u8f93\u51fa\u662f\u90a3\u4e9b\u5bf9\u8bbe\u8ba1\u7ecf\u6d4e\u9ad8\u6548\u7684\u9a8c\u8bc1\u5b9e\u9a8c\u5177\u6709\u9ad8\u6548\u7528\u7684\u8f93\u51fa\uff0c\u662f\u521d\u671f\u5b9e\u9645\u5e94\u7528\u7684\u5173\u952e\u3002\u89e3\u91ca\u6a21\u578b\u9884\u6d4b\u5b58\u5728\u591a\u79cd\u65b9\u6cd5\uff0c\u5305\u62ec\u56e0\u679c\u5efa\u6a21\u3001\u7a00\u758f\u7279\u5f81\u5316\u548c\u53cd\u4e8b\u5b9e\u63a8\u7406\uff0c\u8fd9\u662f\u4e00\u4e2a\u9ad8\u5ea6\u6d3b\u8dc3\u7684\u7814\u7a76\u9886\u57df\u3002\u6784\u5efa\u76f4\u89c2\u7684\u754c\u9762\uff0c\u901a\u8fc7\u5176\u4ed6\u6a21\u578b<\/span>\uff08\u5982 AI \u7814\u7a76\u667a\u80fd\u4f53\uff09<\/span>\u4fc3\u8fdb AIVC \u7684\u7814\u7a76\u548c\u89e3\u91ca\uff0c\u5c06\u8fdb\u4e00\u6b65\u63d0\u9ad8\u4e0b\u6e38\u6548\u7528\u3002<\/span><\/span><\/p>\n

            \n
          • \n

            \u6784\u5efa\u534f\u4f5c\u7ec6\u80de\u5efa\u6a21\u6846\u67b6<\/span><\/span><\/strong><\/p>\n<\/li>\n<\/ul>\n

            AIVC \u7684\u6210\u529f\u5f00\u53d1\u9700\u8981\u8de8\u5b66\u79d1\u5408\u4f5c\u3002\u4e3a\u6b64\uff0c\u6211\u4eec\u5c55\u671b\u672a\u6765\u7684 AIVC \u5e73\u53f0\u5c06\u53d1\u5c55\u6210\u4e3a\u4e00\u4e2a\u5f00\u653e\u3001\u4e92\u8054\u7684\u7efc\u5408\u6027\u67a2\u7ebd\u3002\u4e00\u65b9\u9762\uff0c\u5b83\u5c06\u4f5c\u4e3a\u79d1\u7814\u4e2d\u5fc3\uff0c\u652f\u6301\u7814\u7a76\u4eba\u5458\u534f\u4f5c\u5f00\u53d1\u7ec6\u80de\u6a21\u578b\u5e76\u8fdb\u884c\u5e7f\u6cdb\u90e8\u7f72\uff1b\u53e6\u4e00\u65b9\u9762\uff0c\u5b83\u4e5f\u5c06\u627f\u62c5\u6559\u80b2\u4f7f\u547d\uff0c\u4e3a\u79d1\u7814\u4eba\u5458\u63d0\u4f9b\u4e13\u4e1a\u57f9\u8bad\uff0c\u5e76\u4e3a\u6559\u80b2\u5de5\u4f5c\u8005\u3001\u60a3\u8005\u53ca\u793e\u4f1a\u516c\u4f17\u8bbe\u8ba1\u4e30\u5bcc\u7684\u53c2\u4e0e\u6d3b\u52a8\u3002\u56e0\u6b64\uff0c\u6295\u8d44\u4e8e\u4fc3\u8fdb AIVC \u5f00\u653e\u534f\u4f5c\u5f00\u53d1\u7684\u57fa\u7840\u8bbe\u65bd\u5e94\u5f53\u662f\u9ad8\u4f18\u5148\u7ea7\u4e8b\u9879\u3002<\/span><\/span><\/p>\n

              \n
            • \n

              \u786e\u4fdd AIVC \u4f7f\u6240\u6709\u4eba\u53d7\u76ca\u5e76\u4fc3\u8fdb\u9053\u5fb7\u548c\u8d1f\u8d23\u4efb\u7684\u4f7f\u7528<\/span><\/span><\/strong><\/p>\n<\/li>\n<\/ul>\n

              \u786e\u4fdd AIVC \u7684\u5f00\u53d1\u548c\u5e94\u7528\u7b26\u5408\u4f26\u7406\u6807\u51c6\uff0c\u4fc3\u8fdb\u5176\u8d1f\u8d23\u4efb\u7684\u4f7f\u7528\u81f3\u5173\u91cd\u8981\u3002\u8fd9\u5305\u62ec\u5236\u5b9a\u660e\u786e\u7684\u653f\u7b56\u548c\u6307\u5bfc\u65b9\u9488\uff0c\u4ee5\u786e\u4fdd\u6240\u6709\u76f8\u5173\u65b9\u90fd\u80fd\u5728\u900f\u660e\u548c\u516c\u6b63\u7684\u73af\u5883\u4e2d\u53c2\u4e0e AIVC \u7684\u7814\u7a76\u548c\u5e94\u7528\u3002\u6b64\u5916\uff0c\u5fc5\u987b\u52a0\u5f3a\u5bf9 AIVC \u6280\u672f\u7684\u76d1\u7ba1\uff0c\u4ee5\u9632\u6b62\u6f5c\u5728\u7684\u6ee5\u7528\u548c\u4e0d\u5f53\u4f7f\u7528\uff0c\u786e\u4fdd\u5176\u5728\u79d1\u5b66\u7814\u7a76\u548c\u793e\u4f1a\u5e94\u7528\u4e2d\u7684\u6b63\u5f53\u6027\u548c\u6709\u6548\u6027\u3002<\/span><\/span><\/p>\n

                \n
              • \n

                \u751f\u6210\u53cd\u6620\u4eba\u7c7b\u591a\u6837\u6027\u7684\u5927\u578b\u5f00\u653e\u6570\u636e\u96c6<\/span><\/span><\/strong><\/p>\n<\/li>\n<\/ul>\n

                \u751f\u6210\u53cd\u6620\u4eba\u7c7b\u591a\u6837\u6027\u7684\u5927\u578b\u5f00\u653e\u6570\u636e\u96c6\u662f\u8bad\u7ec3 AIVC \u6a21\u578b\u4e0d\u53ef\u6216\u7f3a\u7684\uff0c\u8fd9\u4e5f\u6784\u6210\u4e86\u4e00\u9879\u91cd\u5927\u6311\u6218\u3002\u5f00\u53d1\u8005\u5728\u6784\u5efaAIVC\u65f6\u5fc5\u987b\u6781\u5176\u8c28\u614e\u5730\u786e\u4fdd\u8fd9\u4e9b\u6570\u636e\u96c6\u5f97\u5230\u9053\u5fb7\u548c\u900f\u660e\u7684\u4f7f\u7528\uff0c\u5e76\u5236\u5b9a\u7b56\u7565\u6765\u964d\u4f4e\u6a21\u578b\u88ab\u865a\u5047\u6570\u636e\u6c61\u67d3\u7684\u98ce\u9669\u3002\u4ece\u4e00\u5f00\u59cb\u5c31\u4e0e\u4f26\u7406\u548c\u76d1\u7ba1\u4e13\u5bb6\u5bc6\u5207\u5408\u4f5c\uff0c\u5bf9\u4e8e\u5efa\u7acb\u65b0\u7684\u76d1\u7ba1\u89c4\u8303\u4ee5\u4fc3\u8fdb AIVC \u7684\u8d1f\u8d23\u4efb\u4f7f\u7528\u81f3\u5173\u91cd\u8981\u3002<\/span><\/span><\/p>\n

                  \n
                • \n

                  \u4e86\u89e3\u4e0d\u540c\u6570\u636e\u7c7b\u578b\u7684\u4ef7\u503c\u4ee5\u4f18\u5148\u8003\u8651\u5927\u89c4\u6a21\u6570\u636e\u751f\u6210<\/span><\/span><\/strong><\/p>\n<\/li>\n<\/ul>\n

                  \u4eba\u5de5\u667a\u80fd\u865a\u62df\u7ec6\u80de\u534f\u4f5c\u5f00\u53d1\u7684\u4e00\u4e2a\u57fa\u672c\u95ee\u9898\u662f\u5e94\u8be5\u6536\u96c6\u54ea\u4e9b\u6570\u636e\u548c\u6a21\u6001\uff0c\u4ee5\u5b9e\u73b0\u8de8\u751f\u7269\u5b66\u80cc\u666f\u548c\u5c3a\u5ea6\u7684\u6cdb\u5316\u3002\u8fd9\u4e9b\u6570\u636e\u9700\u8981\u6db5\u76d6\u4e0d\u540c\u7269\u79cd\u3001\u9886\u57df\u548c\u6a21\u6001\u7684\u751f\u7269\u5b66\u5e7f\u5ea6\uff0c\u4ee3\u8868\u751f\u547d\u7684\u5f02\u8d28\u6027\uff0c\u540c\u65f6\u4fdd\u6301\u8db3\u591f\u7684\u6df1\u5ea6\u4ee5\u533a\u5206\u771f\u5b9e\u4fe1\u53f7\u548c\u566a\u58f0\u3002\u6570\u636e\u751f\u6210\u7684\u4e00\u4e2a\u5173\u952e\u65b9\u9762\u662f\u540c\u65f6\u6d4b\u91cf\u65f6\u95f4\u548c\u7269\u7406\u5c3a\u5ea6\uff0c\u540c\u65f6\u5141\u8bb8\u5bf9\u7cfb\u7edf\u8fdb\u884c\u6270\u52a8\u3002<\/span><\/span><\/p>\n


                  <\/span><\/span><\/p>\n

                  \n
                  \n
                  \n
                  \n
                  \n

                  <\/span><\/section>\n

                  <\/span><\/section>\n
                  \n
                  <\/section>\n<\/section>\n<\/section>\n
                  \n

                  3. AIVCs\u7684\u5173\u952e\u80fd\u529b<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

                  \n

                  <\/span><\/section>\n

                  <\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                  <\/span><\/p>\n

                  \u6211\u4eec\u5bf9AIVC\u7684\u6784\u60f3\u662f\uff0c\u5b83\u662f\u4e00\u4e2a\u80fd\u591f\u5b66\u4e60\u5e76\u6a21\u62df\u7ec6\u80de\u53ca\u7ec6\u80de\u7cfb\u7edf\u5728\u591a\u53d8\u6761\u4ef6\u4e0e\u60c5\u5883\u4e0b\u884c\u4e3a\u7684<\/span><\/span>\u6a21\u62df\u5668 (learned simulator)<\/span><\/span><\/strong> \u3002\u8fd9\u4e9b\u60c5\u5883\u5305\u62ec\u5206\u5316\u72b6\u6001\u3001\u6270\u52a8\u3001\u75be\u75c5\u72b6\u6001\u3001\u968f\u673a\u6ce2\u52a8\u548c\u73af\u5883\u6761\u4ef6<\/span>\uff08\u56fe 1\uff09<\/span>\u3002 \u5b83\u5fc5\u987b\u80fd\u591f\u8de8\u8d8a\u4e0d\u540c\u7684\u751f\u7269\u5b66\u5c3a\u5ea6\u3001\u65f6\u95f4\u7ef4\u5ea6\u548c\u6570\u636e\u6a21\u5f0f\u8fdb\u884c\u5de5\u4f5c\uff0c\u5e76\u6700\u7ec8\u5e2e\u52a9\u63ed\u793a\u7ec6\u80de\u7cfb\u7edf\u7684\u201c\u7f16\u7a0b\u8bed\u8a00\u201d\uff0c\u540c\u65f6\u4e3a\u5de5\u7a0b\u5316\u5e94\u7528\u63d0\u4f9b\u4e00\u4e2a\u53ef\u64cd\u4f5c\u7684\u63a5\u53e3\u3002<\/span><\/span><\/span><\/p>\n

                  \u5177\u4f53\u800c\u8a00\uff0c\u4e00\u4e2a\u7406\u60f3\u7684AIVC\u9700\u5177\u5907\u4ee5\u4e0b\u4e09\u9879\u6838\u5fc3\u80fd\u529b\uff0c\u4ee5\u8d4b\u80fd\u7814\u7a76\u4eba\u5458\uff1a<\/span><\/span><\/p>\n

                  (1) \u521b\u5efa\u901a\u7528\u8868\u793a (Universal Representation, UR)<\/span><\/span><\/strong>\uff1a\u80fd\u591f\u8de8\u8d8a\u7269\u79cd\u3001\u6570\u636e\u6a21\u5f0f\u3001\u6570\u636e\u96c6\u548c\u5177\u4f53\u60c5\u5883<\/span>\uff08\u5305\u62ec\u7ec6\u80de\u7c7b\u578b\u3001\u53d1\u80b2\u9636\u6bb5\u548c\u5916\u90e8\u6761\u4ef6\uff09<\/span>\uff0c\u5efa\u7acb\u751f\u7269\u72b6\u6001\u7684\u901a\u7528\u8868\u793a\u3002<\/span><\/span><\/span><\/p>\n

                  (2) \u9884\u6d4b\u529f\u80fd\u4e0e\u673a\u5236<\/span><\/span><\/strong>\uff1a\u80fd\u591f\u9884\u6d4b\u7ec6\u80de\u7684\u529f\u80fd\u3001\u884c\u4e3a\u548c\u52a8\u6001\u53d8\u5316\uff0c\u5e76\u63ed\u793a\u5176\u80cc\u540e\u7684\u6f5c\u5728\u673a\u5236\u3002<\/span><\/span><\/span><\/p>\n

                  (3) \u6267\u884c\u8ba1\u7b97\u673a\u6a21\u62df\u5b9e\u9a8c (In silico experiments)<\/span><\/span><\/strong>\uff1a\u80fd\u591f\u901a\u8fc7\u6a21\u62df\u5b9e\u9a8c\u6765\u751f\u6210\u548c\u68c0\u9a8c\u65b0\u7684\u79d1\u5b66\u5047\u8bbe\uff0c\u5e76\u6307\u5bfc\u540e\u7eed\u7684\u6570\u636e\u6536\u96c6\uff0c\u4ece\u800c\u9ad8\u6548\u5730\u6269\u5c55\u865a\u62df\u7ec6\u80de\u81ea\u8eab\u7684\u80fd\u529b \u3002<\/span><\/span><\/span><\/p>\n

                  \u63a5\u4e0b\u6765\uff0c\u6211\u4eec\u5c06\u5bf9\u8fd9\u4e9b\u5173\u952e\u80fd\u529b\u8fdb\u884c\u8be6\u7ec6\u9610\u8ff0\u3002<\/span><\/span><\/p>\n

                    \n
                  • \n

                    \u901a\u7528\u8868\u793a<\/span><\/span><\/strong> (Universal representation, UR)<\/span><\/span><\/span><\/p>\n<\/li>\n<\/ul>\n

                    AIVC \u5c06\u751f\u7269\u6570\u636e\u6620\u5c04\u5230\u901a\u7528\u8868\u5f81\u7a7a\u95f4<\/span> (UR spaces)\uff08\u56fe 1A\uff09<\/span>\uff0c\u8fd9\u4e0d\u4ec5\u6709\u52a9\u4e8e\u6d1e\u5bdf\u4e0d\u540c\u751f\u7269\u72b6\u6001\u95f4\u7684\u5171\u4eab\u7279\u5f81\uff0c\u8fd8\u80fd\u4f5c\u4e3a\u4e00\u4e2a\u5168\u9762\u7684\u53c2\u7167\u57fa\u51c6 \u3002\u8fd9\u4e9b UR \u5e94\u80fd\u6574\u5408\u5206\u5b50\u3001\u7ec6\u80de\u548c\u591a\u7ec6\u80de\u8fd9\u4e09\u4e2a\u7269\u7406\u5c3a\u5ea6\uff0c\u5e76\u5bb9\u7eb3\u6765\u81ea\u4efb\u4f55\u76f8\u5173\u6570\u636e\u6a21\u5f0f\u548c\u80cc\u666f\u7684\u4fe1\u606f \u3002\u8fd9\u79cd\u6574\u5408\u80fd\u529b\u4f7f\u7814\u7a76\u4eba\u5458\u53ef\u4ee5\u5728 AIVC \u6846\u67b6\u5185\uff0c\u5229\u7528\u5176\u4e30\u5bcc\u7684\u5b58\u91cf\u751f\u7269\u5b66\u77e5\u8bc6\u6765\u8865\u5145\u65b0\u6570\u636e\uff0c\u4ece\u800c\u5f25\u5408\u4e0d\u540c\u6570\u636e\u96c6\u4e4b\u95f4\u7684\u9e3f\u6c9f\u3002\u81f3\u5173\u91cd\u8981\u7684\u662f\uff0c\u8fd9\u79cd\u591a\u5c42\u6b21\u7684\u8868\u793a\u5e94\u80fd<\/span><\/span>\u6cdb\u5316 (generalize)<\/span><\/span><\/strong> \u5230\u8bad\u7ec3\u6570\u636e\u4e2d\u672a\u66fe\u51fa\u73b0\u8fc7\u7684\u65b0\u72b6\u6001\u3002\u8fd9\u79cd<\/span><\/span>\u6d8c\u73b0\u80fd\u529b (emergent capability)<\/span><\/span><\/strong> \u5c06\u89e3\u9501\u5bf9\u5168\u65b0\u751f\u7269\u5b66\u72b6\u6001\u7684\u53d1\u73b0\uff0c\u8fd9\u4e9b\u72b6\u6001\u53ef\u80fd\u4ece\u672a\u88ab\u76f4\u63a5\u89c2\u5bdf\u5230\uff0c\u751a\u81f3\u5728\u81ea\u7136\u754c\u4e2d\u672c\u4e0d\u5b58\u5728 \u3002\u4f8b\u5982\uff0cAIVC \u901a\u8fc7\u5b66\u4e60\u5de8\u566c\u7ec6\u80de\u7684\u708e\u75c7\u72b6\u6001\uff0c\u6216\u8bb8\u80fd\u9884\u6d4b\u51fa\u5c0f\u80f6\u8d28\u7ec6\u80de\u4e2d\u4e00\u79cd\u524d\u6240\u672a\u77e5\u7684\u708e\u75c7\u72b6\u6001\u3002\u6b64\u5916\uff0cAIVC \u8fd8\u5e94\u80fd\u9884\u6d4b\u7531\u5e72\u9884\u63aa\u65bd<\/span>\uff08\u6216\u4e3a\u8fbe\u5230\u7279\u5b9a\u72b6\u6001\u6240\u9700\u7684\u5e72\u9884\u63aa\u65bd\uff09<\/span>\u6240\u4ea7\u751f\u7684\u65b0\u72b6\u6001\uff0c\u8fd9\u5c06\u4e3a\u7ec6\u80de\u5de5\u7a0b\u548c\u5408\u6210\u751f\u7269\u5b66\u9886\u57df\u7684\u4e0b\u6e38\u5e94\u7528\u63d0\u4f9b\u5e7f\u9614\u7684\u53ef\u80fd\u6027\u3002<\/span><\/span><\/span><\/p>\n

                      \n
                    • \n

                      \u9884\u6d4b\u7ec6\u80de\u884c\u4e3a\u548c\u7406\u89e3\u673a\u5236<\/span><\/span><\/strong> (Predicting cell behavior and understanding mechanisms)<\/span><\/span><\/span><\/p>\n<\/li>\n<\/ul>\n

                      AIVC\u7684\u4e00\u9879\u51b3\u5b9a\u6027\u529f\u80fd<\/span> (defining function) <\/span>\u662f\u5176\u5efa\u6a21\u7ec6\u80de\u54cd\u5e94\u548c\u52a8\u529b\u5b66\u7684\u80fd\u529b \u3002\u901a\u8fc7\u5728\u8986\u76d6\u4e0d\u540c\u60c5\u5883\u548c\u5c3a\u5ea6\u7684\u5927\u91cf\u5feb\u7167\u5f0f\u3001\u65f6\u95f4\u5206\u8fa8\u3001\u5e72\u9884\u6027\u548c\u975e\u5e72\u9884\u6027\u6570\u636e\u96c6\u4e0a\u8fdb\u884c\u8bad\u7ec3\uff0cAIVC \u53ef\u4ee5\u6df1\u5165\u7406\u89e3\u5728\u81ea\u7136\u6216\u5de5\u7a0b\u4fe1\u53f7<\/span>\uff08\u5982\u5316\u5b66\u3001\u9057\u4f20\u6270\u52a8\u53ca\u5176\u7ec4\u5408\uff09<\/span>\u9a71\u52a8\u4e0b\u53d1\u751f\u7684\u5206\u5b50\u3001\u7ec6\u80de\u548c\u7ec4\u7ec7\u5c42\u9762\u7684\u52a8\u529b\u5b66\u8fc7\u7a0b \u3002AIVC\u5e94\u80fd\u9884\u6d4b\u90a3\u4e9b\u5728\u5b9e\u9a8c\u5ba4\u4e2d\u4ece\u672a\u6d4b\u8bd5\u8fc7\u7684\u6270\u52a8\u6240\u4ea7\u751f\u7684\u54cd\u5e94\uff0c\u5e76\u80fd\u8003\u8651\u5230\u6270\u52a8\u53d1\u751f\u65f6\u7279\u5b9a\u7ec6\u80de\u73af\u5883\u7684\u7279\u5f81\u3002<\/span><\/span><\/p>\n

                      \u540c\u65f6\uff0cAIVC \u4e5f\u5e94\u80fd\u6a21\u62df\u7ec6\u80de\u72b6\u6001\u968f\u65f6\u95f4\u7684\u6f14\u53d8\uff0c\u4ee5\u53ca\u7531\u6b64\u4ea7\u751f\u7684\u591a\u7ec6\u80de\u7a7a\u95f4\u5e03\u5c40 \u3002\u901a\u8fc7\u5bf9\u7ec6\u80de\u6574\u4f53\u72b6\u6001\u7684\u77ac\u65f6\u7279\u6027\u548c\u6301\u7eed\u53d8\u5316\u7684\u5efa\u6a21\uff0cAIVC \u53ef\u4ee5\u63ed\u793a\u5728\u53d1\u80b2\u3001\u7a33\u6001\u7ef4\u6301\u3001\u75be\u75c5\u53d1\u751f\u4e0e\u53d1\u5c55\u7b49\u591a\u6837\u5316\u52a8\u6001\u8fc7\u7a0b\u4e2d\uff0c\u6b64\u524d\u672a\u88ab\u7814\u7a76\u8fc7\u7684\u6f14\u5316\u8f68\u8ff9 \u3002<\/span><\/span><\/p>\n

                      \u53e6\u4e00\u4e2a\u5173\u952e\u6311\u6218\u662f\u7406\u89e3\u8fd9\u4e9b\u8868\u578b\u548c\u8f68\u8ff9\u80cc\u540e\u7684\u5206\u5b50\u673a\u5236 \u3002\u901a\u8fc7\u6a21\u62df\u4e0d\u540c\u5e72\u9884\u63aa\u65bd\u7684\u6548\u679c\uff0cAIVC \u53ef\u4ee5\u63d0\u51fa\u6f5c\u5728\u7684\u56e0\u679c\u56e0\u7d20 \u3002\u5c3d\u7ba1\u4ec5\u51ed\u8ba1\u7b97\u4e0d\u4e00\u5b9a\u80fd\u5b8c\u5168\u63ed\u793a\u56e0\u679c\u5173\u7cfb\uff0c\u4f46AIVC \u6709\u6f5c\u529b<\/span><\/span>\u7f29\u5c0f\u53ef\u80fd\u5047\u8bbe\u7684\u7a7a\u95f4 (reduce the space of possible hypotheses)<\/span><\/span><\/strong> \u3002\u5b83\u80fd\u63d0\u51fa\u4e0e\u4e0d\u540c\u8868\u578b\u76f8\u5173\u7684\u6f5c\u5728\u56e0\u679c\u56e0\u5b50\uff0c\u5e76\u7ed9\u51fa\u76f8\u5e94\u7684\u4e0d\u786e\u5b9a\u6027\u7a0b\u5ea6\uff0c\u4ece\u800c\u8ba9\u79d1\u5b66\u5bb6\u80fd\u66f4\u6709\u9488\u5bf9\u6027\u5730\u901a\u8fc7\u5b9e\u9a8c\u8fdb\u884c\u9a8c\u8bc1 \u3002<\/span><\/span><\/span><\/p>\n

                        \n
                      • \n

                        \u8ba1\u7b97\u673a\u6a21\u62df\u5b9e\u9a8c\u548c\u6307\u5bfc\u6570\u636e\u751f\u6210<\/span><\/span><\/strong><\/p>\n<\/li>\n<\/ul>\n

                        \u5728\u73b0\u5b9e\u5e94\u7528\u4e2d\uff0cAIVC \u7684\u53e6\u4e00\u9879\u51b3\u5b9a\u6027\u529f\u80fd\u662f\u5176\u6307\u5bfc\u6570\u636e\u751f\u6210\u548c\u5b9e\u9a8c\u8bbe\u8ba1\u7684\u80fd\u529b \u3002\u7814\u7a76\u4eba\u5458\u5e94\u80fd\u901a\u8fc7\u5f53\u524d\u5b9e\u9a8c\u5ba4\u5b9e\u9a8c\u7684<\/span><\/span>\u8ba1\u7b97\u5b6a\u751f\u4f53 (computational twins)<\/span><\/span><\/strong>\u2014\u2014\u5728\u6b64\u6211\u4eec\u79f0\u4e4b\u4e3a<\/span><\/span>\u865a\u62df\u4eea\u5668 (virtual instruments, VIs)<\/span><\/span><\/strong>\u2014\u2014\u6765\u67e5\u8be2 AIVC \u3002\u4f8b\u5982\uff0c\u865a\u62df\u5b9e\u9a8c\u53ef\u4ee5\u6a21\u62df\u90a3\u4e9b\u96be\u4ee5\u5728\u4f53\u5916\u57f9\u517b\u7684\u7ec6\u80de\u7c7b\u578b\u7684\u5b9e\u9a8c\uff0c\u6216\u8005\u4ece\u4f4e\u6210\u672c\u6d4b\u91cf<\/span>\uff08\u5982\u65e0\u6807\u8bb0\u6210\u50cf\uff09<\/span>\u4e2d\u63a8\u65ad\u51fa\u6602\u8d35\u7684\u8bfb\u6570<\/span>\uff08\u5982\u5355\u7ec6\u80de\u8f6c\u5f55\u7ec4\uff09<\/span> \u3002\u865a\u62df\u5b9e\u9a8c\u8fd8\u80fd\u4ee5\u5b9e\u9a8c\u5ba4\u65e0\u6cd5\u4f01\u53ca\u7684\u89c4\u6a21\u7b5b\u9009\u6d77\u91cf\u7684\u6f5c\u5728\u6270\u52a8\u6e90\uff0c\u8fd9\u5728\u9762\u5bf9\u7ec4\u5408\u6270\u52a8\u7684\u6307\u6570\u7ea7\u641c\u7d22\u7a7a\u95f4\u65f6\uff0c\u663e\u5f97\u5c24\u4e3a\u5b9d\u8d35\u3002<\/span><\/span><\/span><\/p>\n

                        AIVC \u5c06\u5f15\u9886\u4e00\u79cd\u8bbe\u8ba1\u751f\u7269\u5b9e\u9a8c\u7684\u65b0\u8303\u5f0f \u3002\u5728\u6b64\u8303\u5f0f\u4e2d\uff0cAIVC\u4e0d\u4ec5\u80fd\u8bbe\u8ba1\u5b9e\u9a8c\u6765\u9a8c\u8bc1\u7279\u5b9a\u7684\u79d1\u5b66\u5047\u8bbe\uff0c\u8fd8\u80fd\u8bbe\u8ba1\u5b9e\u9a8c\u6765\u589e\u5f3a\u81ea\u8eab\u7684\u80fd\u529b \u3002AIVC\u80fd\u4e3a\u5176\u9884\u6d4b\u5206\u914d\u7f6e\u4fe1\u5ea6\u503c\uff0c\u4ece\u800c\u5b9e\u73b0\u4ea4\u4e92\u5f0f\u67e5\u8be2\uff0c\u5f15\u5bfc\u5b9e\u9a8c\u8005\u4e3a\u4f4e\u7f6e\u4fe1\u5ea6\u533a\u57df\u751f\u6210\u6700\u6709\u6548\u7684\u6570\u636e\u4ee5\u8fdb\u884c\u6a21\u578b\u4f18\u5316 \u3002\u901a\u8fc7\u6269\u5c55\u4e3a\u4e00\u4e2a\u4e3b\u52a8\u3001\u8fed\u4ee3\u7684<\/span><\/span>\u5b9e\u9a8c\u5ba4\u5728\u73af\u201d(lab-in-the-loop)<\/span><\/span><\/strong>\u6d41\u7a0b\uff0c AIVC \u7684\u6027\u80fd\u5c06\u5f97\u5230\u9ad8\u6548\u4e14\u6709\u9488\u5bf9\u6027\u7684\u63d0\u5347 \u3002\u6700\u7ec8\uff0cAIVC\u751a\u81f3\u53ef\u80fd\u80fd\u591f\u8bc6\u522b\u51fa\u81ea\u8eab\u5728\u751f\u7269\u5b66\u7406\u89e3\u4e0a\u7684\u5173\u952e\u7a7a\u767d\uff0c\u5e76\u63d0\u51fa\u586b\u8865\u8fd9\u4e9b\u7a7a\u767d\u7684\u6700\u6709\u6548\u8def\u5f84 \u3002<\/span><\/span><\/span><\/p>\n


                        <\/span><\/h3>\n

                        \n
                        \n
                        \n
                        \n
                        \n

                        <\/span><\/section>\n

                        <\/span><\/section>\n
                        \n
                        <\/section>\n<\/section>\n<\/section>\n
                        \n

                        4. \u6784\u5efaAIVC\u7684\u65b9\u6cd5<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

                        \n

                        <\/span><\/section>\n

                        <\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                        <\/span><\/p>\n

                        \u6211\u4eec\u8bbe\u60f3 AIVC \u662f\u4e00\u4e2a\u7efc\u5408\u7684 AI \u6846\u67b6\uff0c\u7531\u51e0\u4e2a\u76f8\u4e92\u8fde\u63a5\u7684\u57fa\u7840\u6a21\u578b<\/span> (foundation models)<\/span> \u7ec4\u6210\uff0c\u8fd9\u4e9b\u6a21\u578b\u5728\u65e5\u76ca\u590d\u6742\u7684\u7ec4\u7ec7\u5c42\u6b21\u4e0a\u8868\u793a\u52a8\u6001\u751f\u7269\u7cfb\u7edf\u2014\u2014\u4ece\u5206\u5b50\u5230\u7ec6\u80de\u3001\u7ec4\u7ec7\u53ca\u66f4\u9ad8\u5c42\u6b21\u3002<\/span><\/span><\/p>\n

                        \u6211\u4eec\u7684\u65b9\u6cd5\u6709\u4e24\u4e2a\u4e3b\u8981\u7ec4\u6210\u90e8\u5206\uff1a<\/span><\/span><\/p>\n

                          \n
                        1. \n

                          \u901a\u7528\u591a\u6a21\u6001\u591a\u5c3a\u5ea6\u751f\u7269\u72b6\u6001\u8868\u5f81<\/span>\uff08UR\uff09 (a universal multi-modal multi-scale biological state representation)<\/span><\/span><\/p>\n<\/li>\n

                        2. \n

                          \u4e00\u7ec4 VI<\/span> (virtual instruments, VIs) <\/span>\u2014\u2014\u5373\u64cd\u4f5c\u6216\u89e3\u7801\u8fd9\u4e9b\u8868\u5f81\u7684\u795e\u7ecf\u7f51\u7edc\u3002<\/span><\/span><\/p>\n<\/li>\n<\/ol>\n

                          \u5c3d\u7ba1\u53ef\u80fd\u5b58\u5728\u6784\u5efa AIVC \u7684\u5176\u4ed6\u65b9\u6cd5\uff0c\u4f46\u6211\u4eec\u76f8\u4fe1\u8fd9\u79cd\u65b9\u6cd5\u5c06\u63d0\u4f9b\u4e00\u4e2a\u53ef\u4ee5\u534f\u4f5c\u548c\u5f00\u653e\u65b9\u5f0f\u7684\u6269\u5c55\u652f\u67b6\u3002<\/span><\/span><\/p>\n

                          4.1 \u901a\u7528\u8868\u793a\uff08UR\uff09<\/span><\/span><\/h4>\n

                          UR \u662f\u7531 AIVC \u57fa\u7840\u6a21\u578b\u4ea7\u751f\u7684<\/span><\/span>\u5d4c\u5165 (embedding)<\/span><\/span><\/strong>\uff0c\u5b83\u5c06\u9ad8\u7ef4\u3001\u591a\u5c3a\u5ea6\u3001\u591a\u6a21\u6001\u7684\u751f\u7269\u6570\u636e\u8f6c\u6362\u4e3a\u80fd\u4fdd\u7559\u6709\u610f\u4e49\u5173\u7cfb\u548c\u6a21\u5f0f\u7684\u6570\u503c\u8868\u793a\u3002AIVC \u901a\u8fc7\u8868\u793a\u4ee5\u4e0b\u4e09\u4e2a\u4e0d\u540c\u7684\u7269\u7406\u5c3a\u5ea6<\/span> (physical scales) <\/span>\u6765\u6355\u83b7\u7ec6\u80de\u751f\u7269\u5b66\uff1a<\/span><\/span><\/span><\/p>\n

                          \u5206\u5b50\u5c3a\u5ea6 (Molecular scale)<\/span><\/span><\/strong>\uff1a\u5355\u4e2a\u7ec6\u80de\u5185\u53d1\u73b0\u7684\u5206\u5b50\u53ca\u5176\u7ed3\u6784\uff1b<\/span><\/span><\/span><\/p>\n

                          \u7ec6\u80de\u5c3a\u5ea6 (Cellular scale)<\/span><\/span><\/strong>\uff1a\u4f5c\u4e3a\u8fd9\u4e9b\u76f8\u4e92\u4f5c\u7528\u5206\u5b50\u548c\u7ed3\u6784\u7684\u7a7a\u95f4\u96c6\u5408\u7684\u5355\u4e2a\u7ec6\u80de\uff1b<\/span><\/span><\/span><\/p>\n

                          \u7ec4\u7ec7\u5c3a\u5ea6 (Multicellular scale)<\/span><\/span><\/strong>\uff1a\u5355\u4e2a\u7ec6\u80de\u5982\u4f55\u5728\u7ec4\u7ec7\u4e2d\u5f7c\u6b64\u76f8\u4e92\u4f5c\u7528\u4ee5\u53ca\u4e0e\u975e\u7ec6\u80de\u73af\u5883\u76f8\u4e92\u4f5c\u7528\u3002<\/span><\/span><\/span><\/p>\n

                          \u8fd9\u4e09\u4e2a\u5c3a\u5ea6\u4e2d\u7684\u6bcf\u4e00\u4e2a\u90fd\u7531\u4e00\u4e2a\u72ec\u7279\u7684 UR \u8868\u793a\uff0c\u5e76\u5efa\u7acb\u5728\u524d\u4e00\u5c42\u751f\u6210\u7684\u62bd\u8c61\u4e4b\u4e0a\uff0c\u4ece\u800c\u8fde\u63a5\u4e0d\u540c\u7684\u5c3a\u5ea6\u3002<\/span><\/span><\/p>\n

                          4.2 \u865a\u62df\u4eea\u5668\uff08VI\uff09<\/span><\/span><\/h4>\n

                          \u5728 UR \u7684\u80cc\u666f\u4e0b\uff0c<\/span><\/span>VI <\/span><\/span><\/strong>\u662f\u4ee5 UR \u4e3a\u8f93\u5165\u5e76\u4ea7\u751f\u671f\u671b\u8f93\u51fa\u7684\u795e\u7ecf\u7f51\u7edc\u3002\u6211\u4eec\u63cf\u8ff0\u4e86\u4e24\u79cd\u7c7b\u578b\u7684 VI\uff1a<\/span><\/span><\/span><\/p>\n

                            \n
                          • \n

                            \u89e3\u7801\u5668 VI\uff08Decoder VI\uff09<\/span><\/span><\/strong><\/p>\n<\/li>\n

                              \n
                            • \n

                              \u8f93\u5165\uff1aUR<\/span><\/span><\/p>\n<\/li>\n

                            • \n

                              \u8f93\u51fa\uff1a\u4eba\u7c7b\u53ef\u7406\u89e3\u7684\u7ed3\u679c\uff0c\u4f8b\u5982\u7ec6\u80de\u7c7b\u578b\u6807\u7b7e\u6216\u5408\u6210\u663e\u5fae\u955c\u56fe\u50cf\u3002<\/span><\/span><\/p>\n<\/li>\n<\/ul>\n

                            • \n

                              \u64cd\u63a7\u5668 VI\uff08Manipulator VI\uff09<\/span><\/span><\/strong><\/p>\n<\/li>\n

                                \n
                              • \n

                                \u8f93\u5165\uff1aUR<\/span><\/span><\/p>\n<\/li>\n

                              • \n

                                \u8f93\u51fa\uff1a\u53e6\u4e00\u4e2a UR\uff0c\u4f8b\u5982\u6270\u52a8\u540e\u6539\u53d8\u7684\u7ec6\u80de\u72b6\u6001\u3002<\/span><\/span><\/p>\n<\/li>\n<\/ul>\n<\/ul>\n

                                \u7531\u4e8e\u8fd9\u4e9b VI \u5728\u76f8\u540c\u7684\u8868\u793a\u4e0a\u8fd0\u884c\uff0c\u5b83\u4eec\u53ef\u4ee5\u5728\u4e0d\u540c\u7684\u7528\u4f8b\u3001\u5b9e\u9a8c\u548c\u6570\u636e\u96c6\u4e4b\u95f4\u5171\u4eab\u548c\u590d\u7528\u3002  \u56e0\u6b64\uff0c\u6211\u4eec\u8bbe\u60f3\u4efb\u4f55\u79d1\u5b66\u5bb6\u90fd\u80fd\u591f\u5728 UR \u7684\u57fa\u7840\u4e0a\u6784\u5efa VI \u5e76\u4e0e\u793e\u533a\u5171\u4eab\u3002  \u6b64\u5916\uff0c\u6784\u5efa\u4e0e\u771f\u5b9e\u4eea\u5668<\/span>\uff08\u5982\u663e\u5fae\u955c\uff09<\/span>\u9ad8\u5ea6\u76f8\u4f3c\u7684 VI\uff0c\u53ef\u80fd\u4fc3\u8fdb\u7279\u5b9a\u4eea\u5668\u7684\u5b9e\u9a8c\u5ba4\u95ed\u73af\u7cfb\u7edf\u7684\u53d1\u5c55\u3002<\/span><\/span><\/p>\n

                                4.3 \u8de8\u7269\u7406\u5c3a\u5ea6\u6784\u5efa UR<\/span><\/span><\/h4>\n

                                \u6bcf\u4e2a\u8868\u5f81\u666e\u904d\u9002\u7528\u4e8e\u7279\u5b9a\u7c7b\u522b\u7684\u751f\u7269\u5b9e\u4f53\u3002\u8fd9\u79cd\u62bd\u8c61\u4f7f\u865a\u62df\u7ec6\u80de\u80fd\u591f\u5728\u8fd9\u4e2a\u901a\u7528\u6846\u67b6\u5b9e\u73b0\u65e0\u7f1d\u6f14\u8fdb\u4ee5\u53ca\u6574\u5408\u65b0\u6570\u636e\uff0c\u65e0\u8bba\u662f\u6765\u81ea\u65b0\u6a21\u6001\u8fd8\u662f\u6765\u81ea\u5206\u5e03\u5916\u6765\u6e90\u7684\u6570\u636e\u3002<\/span><\/span><\/p>\n

                                <\/section>\n

                                \u56fe2. AIVC \u5c06\u662f\u4e00\u4e2a \u591a\u5c3a\u5ea6\u57fa\u7840\u6a21\u578b\uff0c\u5728\u6bcf\u4e2a\u7269\u7406\u5c3a\u5ea6\u4e0a\u5b66\u4e60\u751f\u7269\u5b9e\u4f53\u7684\u4e0d\u540c\u8868\u5f81\uff08\u56fe 2C\uff09\u3002\u8fd9\u4e9b\u8868\u5f81\u53ef\u4ee5\u805a\u5408\u5728\u4e00\u8d77\u5e76\u8f6c\u6362\uff0c\u4ee5\u4ea7\u751f\u4e0b\u4e00\u4e2a\u66f4\u9ad8\u7269\u7406\u5c3a\u5ea6\u7684\u8868\u5f81\u3002\u5e76\u5f62\u6210\u4e00\u79cd\u9012\u5f52\u67b6\u6784\u6a21\u5f0f\uff0c\u4ece\u5355\u4e2a\u5206\u5b50\u6c34\u5e73\u5e94\u7528\u5230\u6574\u4e2a\u7ec4\u7ec7\u548c\u5668\u5b98\u5c3a\u5ea6\uff0c\u4f7f\u6a21\u578b\u5728\u4e0d\u540c\u751f\u7269\u5c3a\u5ea6\u95f4\u4fdd\u6301\u4e00\u81f4\u6027\uff08\u56fe 2A\uff09\u3002<\/span><\/span><\/p>\n

                                4.4 \u6784\u5efa AIVC \u7684\u4eba\u5de5\u667a\u80fd\u65b9\u6cd5<\/span><\/span><\/h4>\n

                                AIVC \u5c06\u6574\u5408\u591a\u79cd\u591a\u6837\u7684\u795e\u7ecf\u7f51\u7edc\u67b6\u6784\u3002\u5c3d\u7ba1\u8fd9\u4e9b\u67b6\u6784\u5e76\u975e\u4e13\u4e3a\u751f\u7269\u5b66\u5e94\u7528\u8bbe\u8ba1\uff0c\u4f46\u5b83\u4eec\u5728\u4e0e\u7279\u5b9a\u7684\u751f\u7269\u6a21\u5f0f\u548c\u6570\u636e\u7279\u6027\uff08<\/span><\/span>inductive biases\uff0c\u5f52\u7eb3\u504f\u7f6e<\/span><\/span><\/strong>\uff09\u76f8\u5339\u914d\u65f6\uff0c\u5df2\u5c55\u73b0\u51fa\u5404\u81ea\u7684\u4f18\u52bf\u3002\u8bb8\u591a\u60c5\u51b5\u4e0b\uff0c\u8fd9\u4e9b\u67b6\u6784\u53ef\u4ee5\u76f8\u4e92\u66ff\u4ee3\uff0c\u56e0\u6b64\u9700\u8981\u6839\u636e\u5b83\u4eec\u5728\u51c6\u786e\u6027\u3001\u901f\u5ea6\u548c\u901a\u7528\u6027\u65b9\u9762\u7684\u8868\u73b0\u8fdb\u884c\u6743\u8861\u53d6\u820d\u3002\u6b64\u5916\uff0c\u7814\u7a76\u793e\u533a\u6b63\u79ef\u6781\u5f00\u53d1\u4e13\u95e8\u9488\u5bf9<\/span>\uff08\u5927\u578b\uff09<\/span>\u751f\u7269\u6570\u636e\u96c6\u7279\u5f81\u91cf\u8eab\u5b9a\u5236\u7684 AI \u67b6\u6784\u3002<\/span><\/span><\/span><\/p>\n

                                <\/section>\n

                                \u56fe3. \u6784\u5efa AIVC \u7684\u4eba\u5de5\u667a\u80fd\u65b9\u6cd5\u4e4bTransformer<\/span><\/span><\/p>\n

                                  \n
                                • \n

                                  Transformer\u795e\u7ecf\u7f51\u7edc (Transformer neural network)\uff1a<\/span><\/span><\/strong>Transformer \u795e\u7ecf\u7f51\u7edc [30] \u7531\u591a\u4e2a Transformer \u5c42\u7ec4\u6210\uff0c\u6bcf\u4e2a Transformer \u5c42\u90fd\u91c7\u7528\u4e00\u7cfb\u5217\u6807\u8bb0<\/span>\uff08\u79bb\u6563\u7684\u4fe1\u606f\u7247\u6bb5\uff0c\u5982\u5355\u8bcd\u3001RNA \u5206\u5b50\u6216\u57fa\u56e0\u8868\u793a\uff09<\/span>\u4f5c\u4e3a\u8f93\u5165\u2014\u2014\u7b2c\u4e00\u5c42\u7684\u521d\u59cb\u6807\u8bb0\u548c\u540e\u7eed\u5c42\u7684\u524d\u4e00\u5c42\u8f93\u51fa\u3002<\/span><\/span><\/span><\/p>\n

                                  \u5728\u6bcf\u4e00\u5c42\u4e2d\uff0c\u6807\u8bb0\u4f7f\u7528\u81ea\u6ce8\u610f\u529b\u6765\u96c6\u6210\u6765\u81ea\u5176\u4ed6\u6807\u8bb0\u7684\u4e0a\u4e0b\u6587\uff0c\u589e\u5f3a\u81ea\u5df1\u7684\u8868\u793a\uff0c\u7136\u540e\u901a\u8fc7\u524d\u9988\u7f51\u7edc\u8fdb\u884c\u5904\u7406\u3002\u8fd9\u79cd\u67b6\u6784\u4ece\u6839\u672c\u4e0a\u53ea\u9700\u8981\u6807\u8bb0\u96c6\u5408\uff0c\u53ef\u4ee5\u5f88\u597d\u5730\u9002\u5e94\u5404\u79cd\u5e94\u7528\u7a0b\u5e8f\u548c\u7528\u4f8b\u3002<\/span><\/span><\/p>\n

                                  \u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u4f20\u9012\u7ed9 Transformer \u7684\u6807\u8bb0\u96c6\u5408\u6ca1\u6709\u4efb\u4f55\u6392\u5e8f\u3002\u6b64\u5916\uff0c\u81ea\u6ce8\u610f\u529b\u673a\u5236\u662f Transformer \u6210\u529f\u7684\u6838\u5fc3\uff0c\u53ef\u4ee5\u88ab\u89c6\u4e3a\u4e00\u79cd\u5f3a\u70c8\u7684\u751f\u7269\u5f52\u7eb3\u504f\u5dee\u3002\u4f8b\u5982\uff0c\u5728\u901a\u8fc7 scRNA-seq \u68c0\u6d4b\u5230\u7684 RNA \u5206\u5b50\u6765\u8868\u793a\u7ec6\u80de\u65f6\uff0c\u6bcf\u4e2a RNA \u5206\u5b50<\/span>\uff08\u8868\u793a\u4e3a\u6807\u8bb0\uff09<\/span>\u4e0e\u5176\u4ed6 RNA \u5206\u5b50\u76f8\u4e92\u4f5c\u7528\uff0c\u901a\u8fc7\u81ea\u6ce8\u610f\u529b\u6a21\u62df\u57fa\u56e0\u76f8\u4e92\u4f5c\u7528 [30]\u3002\u4f7f\u7528\u57fa\u56e0\u7684\u6570\u5b57\u8868\u793a\u5b9a\u5236\u8f93\u5165\u6807\u8bb0\u8fdb\u4e00\u6b65\u5141\u8bb8\u96c6\u6210\u4ece\u5355\u4e2a\u57fa\u56e0\u5230\u6574\u4e2a\u7ec6\u80de\u7684\u4e0d\u540c\u751f\u7269\u6570\u636e\u5c3a\u5ea6 [59, 60]\u3002<\/span><\/span><\/p>\n

                                  \u6b64\u5916\uff0c\u5c06\u4f4d\u7f6e\u7f16\u7801\u5f15\u5165\u6807\u8bb0\u4f7f Transformer \u80fd\u591f\u901a\u8fc7\u5408\u5e76\u7279\u5b9a\u4e8e\u5e8f\u5217\u7684\u4f9d\u8d56\u5173\u7cfb\u6765\u5904\u7406\u5e8f\u5217<\/span>\uff08\u4f8b\u5982\u81ea\u7136\u8bed\u8a00 [30] \u6216\u751f\u7269\u5e8f\u5217\u5982 DNA [48,142]\uff09<\/span>\u3002\u8fd9\u79cd\u65b9\u6cd5\u5728\u63a9\u7801\u8bed\u8a00\u5efa\u6a21\u7b49\u5e94\u7528\u4e2d\u81f3\u5173\u91cd\u8981\uff0c\u5176\u4e2d\u6a21\u578b\u9884\u6d4b\u5e8f\u5217\u4e2d\u7f3a\u5931\u7684\u6807\u8bb0\uff0c\u4ece\u800c\u589e\u5f3a\u5176\u5bf9\u6570\u636e\u4e2d\u4e0a\u4e0b\u6587\u5173\u7cfb\u7684\u7406\u89e3\u3002<\/span><\/span><\/p>\n

                                  \u521b\u65b0\u4e0d\u65ad\u5b8c\u5584 Transformer\uff0c\u63d0\u9ad8\u5176\u5904\u7406\u66f4\u957f\u5e8f\u5217\u7684\u80fd\u529b\u5e76\u63d0\u9ad8\u6548\u7387\u3002\u4f8b\u5982\uff0c\u72b6\u6001\u7a7a\u95f4\u6a21\u578b\u7b49\u8fdb\u6b65\u80fd\u591f\u751f\u6210\u5e7f\u6cdb\u7684 DNA \u5e8f\u5217 [51]\u3002<\/span><\/span><\/p>\n

                                  <\/section>\n

                                  \u56fe4.\u6784\u5efa AIVC \u7684\u4eba\u5de5\u667a\u80fd\u65b9\u6cd5\u4e4bCNN<\/span><\/span><\/p>\n


                                  <\/span><\/p>\n<\/li>\n

                                • \n

                                  \u5377\u79ef\u795e\u7ecf\u7f51\u7edc (Convolutional Neural Network, CNN)\uff1a<\/span><\/span><\/strong><\/p>\n

                                  CNN \u662f\u4e00\u79cd\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff0c\u4e3b\u8981\u7528\u4e8e\u5206\u6790\u56fe\u50cf [73,74]\u3002\u5b83\u7531\u591a\u4e2a\u5c42\u7ec4\u6210\uff0c\u901a\u8fc7\u53cd\u5411\u4f20\u64ad\u81ea\u52a8\u81ea\u9002\u5e94\u5730\u5b66\u4e60\u7279\u5f81\u7684\u7a7a\u95f4\u5c42\u6b21\u7ed3\u6784\u3002\u5377\u79ef\u5c42\u3001\u6c60\u5316\u5c42\u548c\u5168\u8fde\u63a5\u5c42\u534f\u540c\u4f5c\u7528\uff0c\u4ece\u8f93\u5165\u6570\u636e\u4e2d\u63d0\u53d6\u7279\u5f81\u5e76\u505a\u51fa\u51b3\u7b56\u3002<\/span><\/span><\/p>\n

                                  \u5728\u751f\u7269\u5b66\u9886\u57df\uff0cCNN \u5df2\u88ab\u8bc1\u660e\u5bf9\u4e8e\u6d89\u53ca\u56fe\u50cf\u6570\u636e\u7684\u4efb\u52a1\u5177\u6709\u65e0\u4ef7\u7684\u4ef7\u503c\u3002\u4f8b\u5982\uff1a<\/span><\/span><\/p>\n<\/li>\n

                                    \n
                                  • \n

                                    \u7ec6\u80de\u548c\u7ec4\u7ec7\u7684\u663e\u5fae\u955c\u56fe\u50cf\u6a21\u5f0f\u8bc6\u522b\uff1b<\/span><\/span><\/p>\n<\/li>\n

                                  • \n

                                    \u591a\u91cd\u6210\u50cf<\/span>\uff08\u540c\u65f6\u6807\u8bb0\u548c\u53ef\u89c6\u5316\u591a\u4e2a\u76ee\u6807\uff09<\/span> [143,144]\uff1b<\/span><\/span><\/p>\n<\/li>\n

                                  • \n

                                    \u4e34\u5e8a\u75c5\u7406\u5b66\u4e2d\u7684 H&E \u67d3\u8272\u7ec4\u7ec7\u5207\u7247\u5206\u6790 [145]\uff1b<\/span><\/span><\/p>\n<\/li>\n

                                  • \n

                                    \u6d3b\u7ec6\u80de\u6210\u50cf<\/span> (live-cell imaging) <\/span>\u4e2d\u7528\u4e8e\u8ddf\u8e2a\u7ec6\u80de\u5185\u6216\u5355\u5206\u5b50\u968f\u65f6\u95f4\u7684\u52a8\u6001\u53d8\u5316 [146]\u3002<\/span><\/span><\/p>\n<\/li>\n<\/ul>\n<\/ul>\n

                                    \u9664\u4e86\u4f20\u7edf\u7684\u56fe\u50cf\u5904\u7406\uff0cCNN \u8fd8\u53ef\u7528\u4e8e DNA \u7b49\u5e8f\u5217\u6570\u636e\u7684\u6a21\u5f0f\u8bc6\u522b [147]\u3002\u5c3d\u7ba1 CNN \u5177\u6709\u5e7f\u6cdb\u5b9e\u7528\u6027\uff0c\u4f46\u5728\u4e00\u4e9b\u4efb\u52a1\u4e2d\u6b63\u9010\u6e10\u88ab\u89c6\u89c9 Transformer \u8865\u5145\u6216\u53d6\u4ee3 [72]\u3002<\/span><\/span><\/section>\n
                                    <\/section>\n

                                    \u56fe5.\u6784\u5efa AIVC \u7684\u4eba\u5de5\u667a\u80fd\u65b9\u6cd5\u4e4bDiffusion\u6a21\u578b<\/span><\/span><\/p>\n

                                      \n
                                    • \n

                                      \u6269\u6563\u6a21\u578b (Diffusion models)\uff1a<\/span><\/span><\/strong>\u6269\u6563\u6a21\u578b\u662f\u4e00\u7c7b\u751f\u6210\u5f0f\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff0c\u80fd\u591f\u5728\u591a\u4e2a\u9886\u57df\u751f\u6210\u9ad8\u8d28\u91cf\u3001\u591a\u6837\u5316\u6837\u672c [148]\u3002\u5176\u5de5\u4f5c\u539f\u7406\u662f\uff1a\u9010\u6e10\u5c06\u968f\u673a\u566a\u58f0\u5206\u5e03\u8f6c\u5316\u4e3a\u7ed3\u6784\u5316\u8f93\u51fa<\/span>\uff08\u56fe\u50cf\u3001\u6587\u672c\u3001\u7ec6\u80de\u72b6\u6001\u7b49\uff09<\/span>\u3002<\/span><\/span><\/span><\/p>\n

                                      \u5728\u6269\u6563\u6a21\u578b\u67b6\u6784\u7684\u57fa\u7840\u4e0a\uff0cflow matching \u7b49\u65b9\u6cd5\u53ef\u4ee5\u6a21\u62df\u968f\u65f6\u95f4\u6f14\u53d8\u7684\u5206\u5e03 [149]\uff0c\u4f7f\u5176\u7279\u522b\u9002\u5408\u4e8e\u52a8\u6001\u53d8\u5316\u548c\u65f6\u95f4\u8fdb\u5c55\u81f3\u5173\u91cd\u8981\u7684\u751f\u7269\u5b66\u5e94\u7528\u3002<\/span><\/span><\/p>\n

                                      \u4f8b\u5982\uff0c\u5b83\u4eec\u80fd\u591f\u751f\u6210\u7ec6\u80de\u5728\u65f6\u95f4\u548c\u7a7a\u95f4\u4e0a\u7684\u53d1\u80b2\u8f68\u8ff9\uff0c\u6216\u9884\u6d4b\u751f\u7269\u7cfb\u7edf\u5bf9\u6cbb\u7597\u7684\u53cd\u5e94 [87]\u3002\u6269\u6563\u4e0e flow matching \u6a21\u578b\u7ed3\u5408\uff0c\u4e3a\u9ad8\u7ef4\u590d\u6742\u751f\u7269\u6570\u636e\u7684\u5efa\u6a21\u63d0\u4f9b\u4e86\u5f3a\u5927\u5de5\u5177\u3002<\/span><\/span><\/p>\n<\/li>\n<\/ul>\n

                                      <\/section>\n

                                      \u56fe6.\u6784\u5efa AIVC \u7684\u4eba\u5de5\u667a\u80fd\u65b9\u6cd5\u4e4bGNN<\/span><\/span><\/p>\n

                                        \n
                                      • \n

                                        \u56fe\u795e\u7ecf\u7f51\u7edc (Graph Neural Network, GNN)\uff1a<\/span><\/span><\/strong>\u662f\u4e00\u7c7b\u53ef\u4ee5\u5bf9\u56fe\u6570\u636e\u8fdb\u884c\u5efa\u6a21\u7684\u67b6\u6784 [150]\u3002\u56fe\u7531\u8282\u70b9\u548c\u8fb9\u7ec4\u6210\uff0c\u662f\u8bb8\u591a\u751f\u7269\u7cfb\u7edf\u7684\u81ea\u7136\u8868\u793a\uff1a<\/span><\/span><\/span><\/p>\n<\/li>\n

                                          \n
                                        • \n

                                          \u86cb\u767d\u8d28\u7ed3\u6784\uff1a\u6b8b\u57fa\u4f5c\u4e3a\u8282\u70b9\uff0c\u5316\u5b66\u952e\u4f5c\u4e3a\u8fb9 [151]\uff1b<\/span><\/span><\/p>\n<\/li>\n

                                        • \n

                                          \u7ec4\u7ec7\u7ed3\u6784\uff1a\u7ec6\u80de\u4f5c\u4e3a\u8282\u70b9\uff0c\u76f8\u90bb\u7ec6\u80de\u901a\u8fc7\u8fb9\u8868\u793a\u7269\u7406\u6216\u5316\u5b66\u8fde\u63a5 [152, 153]\u3002<\/span><\/span><\/p>\n<\/li>\n<\/ul>\n<\/ul>\n

                                          GNN \u53ef\u7528\u4e8e\u8282\u70b9\u3001\u8fb9\u6216\u6574\u4e2a\u56fe\u7684\u9884\u6d4b [154]\u3002\u5728\u8282\u70b9\u7ea7 GNN \u4e2d\uff0c\u6bcf\u4e00\u5c42\u4e2d\u8282\u70b9\u4f1a\u6574\u5408\u90bb\u5c45\u7684\u7279\u5f81\u4fe1\u606f<\/span>\uff08\u6d88\u606f\u4f20\u9012\u673a\u5236\uff09<\/span>\uff0c\u5e76\u9010\u5c42\u6269\u5c55\u611f\u53d7\u91ce\u3002\u5728\u751f\u7269\u5e94\u7528\u4e2d\uff0c\u4f8b\u5982\u5728\u7a7a\u95f4\u8f6c\u5f55\u7ec4\u6570\u636e\u4e2d\uff0cGNN \u53ef\u4ee5\u5c06\u6bcf\u4e2a\u7ec6\u80de\u7684\u57fa\u56e0\u8868\u8fbe\u4f5c\u4e3a\u521d\u59cb\u7279\u5f81\uff0c\u5e76\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u901a\u8fc7\u90bb\u5c45\u4fe1\u606f\u66f4\u65b0\uff0c\u4ece\u800c\u63ed\u793a\u7ec6\u80de\u95f4\u7684\u7a7a\u95f4\u76f8\u4e92\u4f5c\u7528\u4e0e\u751f\u6001\u4f4d [153]\u3002<\/span><\/span><\/h3>\n


                                          <\/span><\/h3>\n

                                          \n
                                          \n
                                          \n
                                          \n
                                          \n

                                          <\/span><\/section>\n

                                          <\/span><\/section>\n
                                          \n
                                          <\/section>\n<\/section>\n<\/section>\n
                                          \n

                                          5. \u6784\u5efaAIVC\u7684\u6570\u636e\u9700\u6c42\u548c\u8981\u6c42<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

                                          \n

                                          <\/span><\/section>\n

                                          <\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                                          <\/span><\/p>\n

                                          AIVC \u7684\u4e00\u4e2a\u5173\u952e\u8003\u8651\u56e0\u7d20\u662f\u5fc5\u987b\u6536\u96c6\u54ea\u4e9b\u6570\u636e\u96c6\u548c\u6a21\u6001\u6765\u5b9e\u73b0\u5176\u6709\u6548\u6784\u5efa\u3002\u4e0e\u4f20\u7edf\u7684\u5b9e\u9a8c\u8bbe\u8ba1\u4e0d\u540c\uff0c\u4f20\u7edf\u5b9e\u9a8c\u8bbe\u8ba1\u4e2d\u6570\u636e\u662f\u4e3a\u4e86\u6d4b\u8bd5\u7279\u5b9a\u7684\u79d1\u5b66\u5047\u8bbe\u800c\u751f\u6210\u7684\uff0c\u800c\u7528\u4e8e\u8bad\u7ec3 AIVC \u7684\u6570\u636e\u6536\u96c6\u5e94\u8be5\u4e13\u6ce8\u4e8e\u786e\u4fdd AIVC \u6240\u671f\u671b\u7684\u5e7f\u6cdb\u9002\u7528\u6027\u548c\u901a\u7528\u6027\u3002\u4e3a\u4e86\u5b9e\u73b0\u8fd9\u4e9b\u76ee\u6807\uff0c\u6570\u636e\u7406\u60f3\u60c5\u51b5\u4e0b\u5e94\u8be5\u8de8\u8d8a\u4e0d\u540c\u7684\u9886\u57df\u548c\u6a21\u6001\uff0c\u5305\u542b\u751f\u7269\u53d8\u5f02\u6027\u7684\u5f02\u8d28\u6027\u548c\u591a\u6837\u6027\uff0c\u5e76\u4f7f\u6a21\u578b\u80fd\u591f\u533a\u5206\u6280\u672f\u566a\u58f0\u3001\u751f\u7269\u53d8\u5f02\u548c\u751f\u7406\u5dee\u5f02\u3002<\/span><\/span><\/p>\n

                                          \u6570\u636e\u751f\u6210\u9700\u8981\u540c\u65f6\u63a2\u7d22\u65f6\u95f4\u548c\u7269\u7406\u5c3a\u5ea6\uff0c\u540c\u65f6\u5141\u8bb8\u7cfb\u7edf\u6270\u52a8\u3002\u5728\u8fd9\u91cc\uff0c\u7ecf\u5178\u6210\u50cf\u6280\u672f [65, 100, 101] \u5305\u62ec\u6d3b\u7ec6\u80de\u6210\u50cf\uff0c\u4ee5\u53ca\u66f4\u65b0\u7684\u7ed3\u6784\u6210\u50cf\u6280\u672f\uff0c\u5982<\/span><\/span>\u51b7\u51bb\u7535\u5b50\u65ad\u5c42\u626b\u63cf (cryoelectron tomography)<\/span><\/span><\/strong> \u548c<\/span><\/span>\u8f6fX\u5c04\u7ebf\u65ad\u5c42\u626b\u63cf (soft X-ray tomography)<\/span><\/span><\/strong>  [66, 102, 103]\uff0c\u4ee5\u53ca\u65b0\u9896\u7684\u7a7a\u95f4\u7ec4\u5b66\u6280\u672f<\/span> (spatial omics technologies) <\/span> [104, 105]\uff0c\u4e3a\u8de8\u5c3a\u5ea6\u5efa\u6a21\u751f\u7269\u5206\u5b50\u548c\u529f\u80fd\u63d0\u4f9b\u4e86\u673a\u4f1a\u3002\u6b64\u5916\uff0c\u751f\u7269\u8fc7\u7a0b\u8de8\u8d8a\u4e86\u5de8\u5927\u7684\u65f6\u95f4\u5c3a\u5ea6\u8303\u56f4\uff0c\u4ece\u76ae\u79d2\u7ea7\u7684\u6700\u5feb\u53cd\u5e94\u5230\u6301\u7eed\u6570\u5c0f\u65f6\u81f3\u4e00\u5929\u7684\u7ec6\u80de\u5206\u88c2\uff0c\u518d\u5230\u5386\u65f6\u6570\u5e74\u7684\u80bf\u7624\u53d1\u5c55\uff0c\u4ee5\u53ca\u6301\u7eed\u6570\u5341\u5e74\u7684\u795e\u7ecf\u9000\u884c\u6027\u53d8\u5316\u3002\u6700\u8fd1\u6784\u5efa\u7684\u901a\u7528\u7ec6\u80de\u56fe\u8c31 [101, 106] \u53ef\u80fd\u6210\u4e3a\u5efa\u6a21\u8f83\u957f\u65f6\u95f4\u5c3a\u5ea6\u4e0a\u7ec6\u80de\u884c\u4e3a\uff08\u5982\u7ec4\u7ec7\u5f62\u6210\uff09\u7684\u5f3a\u5927\u8d44\u6e90\u3002\u9700\u8981\u65b0\u7684\u65b9\u6cd5\u6765\u6784\u5efa\u53ef\u6bd4\u8f83\u7684\u6570\u636e\u96c6\uff0c\u4ee5\u6355\u83b7\u7ec6\u80de\u5728\u8f83\u77ed\u65f6\u95f4\u5c3a\u5ea6\u4e0a\u7684\u884c\u4e3a\uff0c\u4f8b\u5982\u901a\u8fc7\u6d3b\u7ec6\u80de\u6210\u50cf\u7b49\u65b9\u6cd5\u3002<\/span><\/span><\/span><\/p>\n

                                          \u9664\u4e86\u5206\u5b50\u6d4b\u91cf\u5916\uff0c\u6570\u636e\u6536\u96c6\u7684\u4e00\u4e2a\u91cd\u8981\u65b9\u9762\u5c06\u662f\u6d4b\u91cf\u751f\u7269\u7269\u7406\u548c\u751f\u7269\u5316\u5b66\u7ec6\u80de\u7279\u6027\uff0c\u4e3a AIVC \u63d0\u4f9b\u7269\u7406\u548c\u5316\u5b66\u73b0\u5b9e\u4e3b\u4e49\u7684\u8fb9\u754c\u3002<\/span><\/span><\/p>\n

                                          AIVC \u53d1\u5c55\u7684\u53e6\u4e00\u4e2a\u91cd\u8981\u9a71\u52a8\u529b\u5c06\u662f<\/span><\/span>\u591a\u6a21\u6001\u6570\u636e\u96c6 (multi-modal datasets)<\/span><\/span><\/strong> \u3002\u4f8b\u5982\uff0c\u8fde\u63a5\u5206\u5b50\u548c\u7a7a\u95f4\u5c3a\u5ea6\u7684\u6570\u636e\u96c6\uff0c\u5982\u5355\u7ec6\u80de\u8f6c\u5f55\u7ec4\u5b66\u6570\u636e\u7ed3\u5408\u7ec4\u7ec7\u5b66\uff0c\u4ee5\u4e86\u89e3\u7ec6\u80de\u5982\u4f55\u76f8\u4e92\u4f5c\u7528\u4ee5\u53ca\u54ea\u4e9b\u5206\u5b50\u7279\u5f81\u652f\u6491\u7740\u4e13\u95e8\u7a7a\u95f4\u751f\u6001\u4f4d\u7684\u5f62\u6210 [107]\u3002\u9700\u8981\u8fdb\u4e00\u6b65\u7684\u6280\u672f\u53d1\u5c55\u6765\u6536\u96c6\u591a\u6a21\u6001\u6570\u636e\uff0c\u4ee5\u6355\u6349\u5206\u5b50\u7279\u5f81\u3001\u7ec6\u80de\u884c\u4e3a\u3001\u7ec6\u80de\u8c03\u63a7\u548c\u7ec4\u7ec7\u4e4b\u95f4\u7684\u5173\u7cfb\u3002<\/span><\/span><\/span><\/p>\n

                                          \u5c3d\u7ba1\u865a\u62df\u7ec6\u80de\u5efa\u6a21\u7684\u6838\u5fc3\u5174\u8da3\u5c06\u96c6\u4e2d\u5728\u4eba\u7c7b\u6570\u636e\u96c6\u4e0a\uff0c\u76ee\u7684\u662f\u7406\u89e3\u75be\u75c5\u5e76\u5e2e\u52a9\u5f00\u53d1\u65b0\u578b\u6cbb\u7597\u65b9\u6cd5\uff0c\u4f46\u4eba\u7c7b\u6570\u636e\u96c6\u5728\u6211\u4eec\u8fdb\u884c\u4f53\u5185\u5bf9\u7167\u5b9e\u9a8c\u548c\u6270\u52a8\u7684\u80fd\u529b\u65b9\u9762\u662f\u6709\u9650\u7684\u3002<\/span><\/span><\/p>\n

                                          \u5728\u6b64\uff0c3D \u7ec4\u7ec7\u751f\u7269\u5b66\u9886\u57df<\/span>\uff08\u5305\u62ec\u7c7b\u5668\u5b98\u7b49\u57f9\u517b\u7cfb\u7edf\uff09<\/span>\u6b63\u9010\u6e10\u5174\u8d77\uff0c\u6210\u4e3a\u4e00\u79cd\u5728 3D \u73af\u5883\u4e2d\u7814\u7a76\u590d\u6742\u7ec4\u7ec7\u7ed3\u6784\u548c\u529f\u80fd[108] \u7684\u5de5\u5177\u3002\u5b83\u5728 3D \u73af\u5883\u4e2d\u8fdb\u884c\u7814\u7a76\uff0c\u540c\u65f6\u5141\u8bb8\u5bf9\u7cfb\u7edf\u8fdb\u884c\u6270\u52a8\u3002\u53e6\u4e00\u4e2a\u8d85\u8d8a\u8fd9\u4e00\u9650\u5236\u7684\u5173\u952e\u9014\u5f84\u5c06\u662f\u5bf9\u6f14\u5316\u5386\u53f2\u7684\u7269\u79cd\u8fdb\u884c\u591a\u6837\u5316\u7684\u5168\u673a\u4f53\u8c31\u5206\u6790\uff0c\u6db5\u76d6\u5404\u79cd\u6270\u52a8\u548c\u4e0d\u540c\u6761\u4ef6 [109, 110, 111]\u3002\u7406\u60f3\u60c5\u51b5\u4e0b\uff0c\u5927\u578b\u6570\u636e\u96c6\u53ef\u4ee5\u5728\u6240\u6709\u4e09\u4e2a\u7269\u7406\u5c3a\u5ea6\u4e0a\u88ab\u6536\u96c6\uff0c\u8fd9\u5c06\u4f7f AIVC \u7684\u5e94\u7528\u80fd\u591f\u4ece\u75be\u75c5\u7814\u7a76\u6269\u5c55\u5230\u5de5\u4e1a\u751f\u7269\u6280\u672f\u3001\u519c\u4e1a\u751f\u7269\u6280\u672f\u3001\u4f20\u67d3\u75c5\u548c\u6c14\u5019\u53d8\u5316\u7b49\u66f4\u5e7f\u9614\u7684\u9886\u57df\u3002\u7136\u800c\uff0c\u57fa\u4e8e\u76ee\u524d\u5728\u7ec6\u80de\u548c\u591a\u7ec6\u80de\u5c3a\u5ea6\u4e0a\u7684\u6570\u636e\u6536\u96c6\u8d8b\u52bf\uff0c\u5bf9\u52a8\u7269\u7ec6\u80de\u8fdb\u884c\u5efa\u6a21\u4ecd\u7136\u662f\u6700\u73b0\u5b9e\u7684\u9009\u62e9\u3002<\/span><\/span><\/p>\n

                                          \u6700\u540e\uff0c\u751f\u7269\u6570\u636e\u751f\u6210\u7684\u4e00\u4e2a\u5173\u952e\u65b9\u9762\u5c06\u662f\u5bf9<\/span><\/span>\u7ec4\u5408\u7a7a\u95f4 (combinatorial spaces)<\/span><\/span><\/strong>\u7684\u63a2\u7d22\uff1a\u751f\u7269\u7a7a\u95f4\u901a\u5e38\u662f\u9ad8\u7ef4\u7684\uff0c\u679a\u4e3e\u5176\u53d8\u4f53\u5728\u4e00\u822c\u60c5\u51b5\u4e0b\u662f\u96be\u4ee5\u5904\u7406\u7684\uff0c\u4f8b\u5982\uff0c\u5f53\u8003\u8651\u57fa\u56e0\u7ec4\u7684\u6240\u6709\u53ef\u80fd\u53d8\u4f53\u65f6\u3002\u5373\u4f7f\u662f\u5c11\u6570\u5b9e\u4f53\u7684\u7ec4\u5408\uff0c\u5982\u679a\u4e3e\u6270\u52a8\u7684\u6210\u5bf9\u6216\u6210\u7ec4\u60c5\u51b5 [47, 90]\uff0c\u5b9e\u9a8c\u8bbe\u8ba1\u4e5f\u53d8\u5f97\u6781\u5176\u5177\u6709\u6311\u6218\u6027\u3002\u9274\u4e8e\u7ec4\u5408\u7684\u53ef\u80fd\u6027\u4f1a\u8fc5\u901f\u81a8\u80c0\uff0c\u8fdc\u8d85\u5b9e\u9a8c\u6216\u8ba1\u7b97\u7684\u53ef\u884c\u8303\u56f4\uff0c\u56e0\u6b64\u5fc5\u987b\u5f00\u53d1\u63a2\u7d22\u8fd9\u4e9b\u53ef\u80fd\u6027\u65b0\u65b9\u6cd5\u3002<\/span><\/span><\/span><\/p>\n

                                            \n
                                          • \n

                                            \u6784\u5efa AIVC \u9700\u8981\u591a\u5c11\u6570\u636e\uff1f<\/span><\/span><\/strong><\/p>\n<\/li>\n<\/ul>\n

                                            \u539f\u59cb\u751f\u7269\u6570\u636e\u7684\u89c4\u6a21\u662f\u4e0d\u53ef\u5426\u8ba4\u7684\uff0c\u4f46\u5373\u4f7f\u662f\u4e00\u4e2a\u4eba\u4f53\u7ec6\u80de\u7cfb\u7edf\u7684\u7eaf\u7cb9\u6807\u79f0\u89c4\u6a21\u4e5f\u540c\u6837\u5e9e\u5927\uff0c\u8fd9\u4f7f\u5f97\u7b2c\u4e00\u6027\u539f\u7406\u4f30\u7b97\u53d8\u5f97\u56f0\u96be\u3002\u4f8b\u5982\uff0c\u751f\u7269\u5e8f\u5217\u6570\u636e\u7684\u77ed\u8bfb\u957f\u5b58\u6863\u62e5\u6709\u8d85\u8fc7 14 petabytes\u7684\u4fe1\u606f [112]\uff0c\u8fd9\u6bd4\u8bad\u7ec3 ChatGPT \u6240\u7528\u7684\u6570\u636e\u96c6\u5927 1000 \u591a\u500d [113]\u3002\u5982\u679c\u7528\u4e8e\u8bad\u7ec3\uff0c\u8fd9\u4e9b\u6570\u636e\u7684\u5f88\u5927\u4e00\u90e8\u5206\u53ef\u80fd\u662f\u5197\u4f59\u7684\u6216\u5177\u6709\u9012\u51cf\u6536\u76ca\uff0c\u56e0\u6b64\u5fc5\u987b\u5f7b\u5e95\u7814\u7a76\u6a21\u578b\u6027\u80fd\u7684\u89c4\u6a21\u6cd5\u5219\u3002<\/span><\/span><\/p>\n

                                            \u9664\u4e86\u6570\u636e\u89c4\u6a21\u5916\uff0c\u6570\u636e\u7684\u591a\u6837\u6027\u548c\u8d28\u91cf\u5bf9\u786e\u4fdd\u6a21\u578b\u6027\u80fd\u81f3\u5173\u91cd\u8981 [114]\u3002\u6765\u81ea\u4eba\u7c7b\u548c\u6a21\u5f0f\u751f\u7269<\/span>\uff08\u5982\u5c0f\u9f20\u548c\u5927\u80a0\u6746\u83cc\uff09<\/span>\u7684\u6570\u636e\u5728\u5e8f\u5217\u548c\u6587\u732e\u6570\u636e\u5e93\u4e2d\u7684\u4ee3\u8868\u6027\u4e0d\u5747\u7b49\uff0c\u5f53\u7528\u4e8e\u8bad\u7ec3\u65f6\uff0c\u4f1a\u7f16\u7801\u5f3a\u70c8\u7684\u7269\u79cd\u504f\u5dee [114]\u3002\u5176\u4ed6\u504f\u5dee\uff0c\u4f8b\u5982\u5728\u6027\u522b\u3001\u7279\u5b9a\u75be\u75c5\u6216\u4eba\u7c7b\u7956\u5148\u7fa4\u4f53\u65b9\u9762\u7684\u504f\u5dee\uff0c\u4e5f\u53ef\u80fd\u964d\u4f4e AIVC \u6a21\u578b\u7684\u5f71\u54cd\u529b [115]\u3002<\/span><\/span><\/p>\n

                                            \u5c3d\u7ba1\u6570\u636e\u65b9\u9762\u9700\u8981\u4ed8\u51fa\u52aa\u529b\uff0c\u4f46\u9a71\u52a8 AIVC \u7684 AI \u6a21\u578b\u5fc5\u987b\u8bbe\u8ba1\u5f97\u80fd\u591f\u627f\u53d7\u5e76\u9002\u5e94\u8fd9\u4e9b\u6311\u6218\uff0c\u5373\u5728\u6574\u5408\u5404\u79cd\u6765\u6e90\u548c\u8d28\u91cf\u7684\u6570\u636e\u96c6\u7684\u80fd\u529b\u65b9\u9762\u8868\u73b0\u51fa\u7a33\u5065\u6027\u3002\u8003\u8651\u5230\u5b9e\u9a8c\u5ba4\u6280\u672f\u7684\u5feb\u901f\u53d1\u5c55<\/span>\uff08\u8fd9\u4f7f\u5f97\u65e0\u6cd5\u5728\u5355\u4e00\u5e73\u53f0\u4e0a\u5b9e\u73b0\u6807\u51c6\u5316\uff09<\/span>\u4ee5\u53ca AIVC \u5fc5\u987b\u6db5\u76d6\u7684\u6a21\u5f0f\u548c\u7ec6\u80de\u7cfb\u7edf\u7684\u5e7f\u6cdb\u591a\u6837\u6027\uff0c\u8fd9\u4e00\u70b9\u81f3\u5173\u91cd\u8981\u3002\u968f\u7740\u865a\u62df\u7ec6\u80de\u5de5\u4f5c\u7684\u6210\u719f\uff0c\u5f00\u53d1\u6a21\u578b\u7684\u79d1\u5b66\u5bb6\u3001\u751f\u6210\u5b9e\u9a8c\u6570\u636e\u7684\u79d1\u5b66\u5bb6\u548c\u8d44\u52a9\u673a\u6784\u4e4b\u95f4\u7684\u5bf9\u8bdd\u5fc5\u987b\u8fdb\u4e00\u6b65\u52a0\u5f3a\u3002<\/span><\/span><\/p>\n


                                            <\/span><\/h3>\n

                                            \n
                                            \n
                                            \n
                                            \n
                                            \n

                                            <\/span><\/section>\n

                                            <\/span><\/section>\n
                                            \n
                                            <\/section>\n<\/section>\n<\/section>\n
                                            \n

                                            6. AIVC\u7684\u6a21\u578b\u8bc4\u4f30\u548c\u53ef\u89e3\u91ca\u6027<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

                                            \n

                                            <\/span><\/section>\n

                                            <\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                                            <\/span><\/p>\n

                                            \u5bf9\u4e8e AIVC \u53d1\u5c55\u800c\u8a00\uff0c\u66f4\u91cd\u8981\u7684\u95ee\u9898\u53ef\u80fd\u4e0d\u662f “\u6211\u4eec\u5982\u4f55\u6784\u5efa\u5b83\u4eec\uff1f” \u800c\u662f “\u6211\u4eec\u5982\u4f55\u5efa\u7acb\u5bf9\u5176\u80fd\u529b\u548c\u51c6\u786e\u6027\u7684\u4fe1\u4efb\uff1f” \u4e3a\u6b64\uff0c\u9700\u8981\u4e00\u4e2a\u5168\u9762\u4e14\u9002\u5e94\u6027\u5f3a\u7684<\/span><\/span>\u57fa\u51c6\u6d4b\u8bd5\u6846\u67b6(benchmarking framework)<\/span><\/span><\/strong>\u3002AIVC \u9700\u8981\u5728\u4f17\u591a\u751f\u7269\u5b66\u80cc\u4efb\u52a1\u4e2d\u5c55\u73b0<\/span><\/span>\u901a\u7528\u6027 (generalizability)<\/span><\/span><\/strong> \u3002\u5b83\u5fc5\u987b\u8003\u8651\u7531\u4e8e\u73af\u5883\u53d8\u5316\u3001\u611f\u67d3\u3001\u9057\u4f20\u53d8\u5f02\u4ee5\u53ca\u5176\u4ed6\u5f15\u8d77\u5206\u5e03\u504f\u79fb<\/span> (distribution shifts)<\/span> \u7684\u56e0\u7d20\u800c\u4e0d\u65ad\u6f14\u53d8\u7684\u52a8\u6001\u5206\u5e03\u3002 [116]\u3002<\/span><\/span><\/span><\/p>\n

                                            \u5373\u4f7f\u5728\u53ef\u63a8\u5e7f\u6027\u4e4b\u5916\uff0c\u65b0\u5174\u80fd\u529b\uff08\u6bd4\u5982\u90a3\u4e9b\u4e0e<\/span><\/span>\u5927\u8bed\u8a00\u6a21\u578b (Large Language Models, LLMs)<\/span><\/span><\/strong> \u76f8\u5173\u7684\u80fd\u529b\uff09\u4e5f\u53ef\u80fd\u4f7f AIVC \u6a21\u578b\u80fd\u591f\u5916\u63a8\u5230\u771f\u6b63\u7684\u5206\u5e03\u5916\u6570\u636e\u3002\u5728\u751f\u7269\u5b66\u80cc\u666f\u4e0b\uff0c\u5728\u8bc4\u4f30\u8fc7\u7a0b\u4e2d\u53ef\u80fd\u5f88\u96be\u786e\u5b9a\u8fd9\u4e2a\u8fb9\u754c\u662f\u5982\u4f55\u5b9a\u4e49\u7684\u3002\u65b0\u5206\u5b50\u3001\u65b0\u7ec6\u80de\u72b6\u6001\uff0c\u751a\u81f3\u65b0\u7269\u79cd\u90fd\u53ef\u80fd\u88ab\u8ba4\u4e3a\u5728\u8bad\u7ec3\u5206\u5e03\u5185\u3002\u56e0\u6b64\uff0c\u5bf9\u65b0\u6570\u636e\u7684\u5916\u63a8\u53ef\u80fd\u4ec5\u9650\u4e8e\u8003\u8651\u8bbe\u8ba1\u81ea\u7136\u754c\u4e2d\u4e0d\u5b58\u5728\u7684\u751f\u7269\u5b9e\u4f53\u3002<\/span><\/span><\/span><\/p>\n

                                            \u8fd9\u79cd\u8bc4\u4f30\u7c7b\u578b\u5df2\u7ecf\u5728\u5206\u5b50\u8bbe\u8ba1\u9886\u57df\u5f97\u5230\u4e86\u8003\u8651\uff0c\u56e0\u4e3a\u8bed\u8a00\u6a21\u578b\u521b\u5efa\u7684\u86cb\u767d\u8d28\uff0c\u5982 esmGFP [52] \u6216 OpenCrispr1 [53]\uff0c\u7a81\u51fa\u4e86\u5b83\u4eec\u4e0e\u4efb\u4f55\u5929\u7136\u5b58\u5728\u7684\u5bf9\u5e94\u7269\u4e4b\u95f4\u7684\u5dee\u5f02\u3002\u5982\u679c\u5728\u8bbe\u8ba1\u8fd9\u4e9b\u6a21\u578b\u65f6\u4ee5\u5916\u63a8\u4e3a\u76ee\u6807\uff0c\u90a3\u4e48\u53ef\u80fd\u9700\u8981\u4f7f\u7528\u751f\u7269\u529b\u5b66\u3001\u57fa\u4e8e\u7269\u7406\u6216\u673a\u5236\u5efa\u6a21\u7684\u989d\u5916\u5f52\u7eb3\u504f\u7f6e\u3001\u5fae\u8c03\u6216\u504f\u597d\u4f18\u5316\u3002<\/span><\/span><\/p>\n

                                            \u5bf9 AIVC \u7684\u8bc4\u4f30\u5e94\u540c\u65f6\u4f18\u5148\u8003\u8651\u5176\u6cdb\u5316\u80fd\u529b\u548c\u53d1\u73b0\u65b0\u751f\u7269\u5b66\u89c4\u5f8b\u7684\u80fd\u529b\u3002\u6cdb\u5316\u80fd\u529b\u8861\u91cf\u4e86\u6a21\u578b\u5728\u672a\u66fe\u89c1\u8fc7\u7684\u8bed\u5883\u4e2d\u7684\u8868\u73b0\uff0c\u4f8b\u5982\u9762\u5bf9\u65b0\u578b\u7ec6\u80de\u7c7b\u578b\u6216\u4e0d\u540c\u7684\u9057\u4f20\u80cc\u666f\u3002\u53ef\u4ee5\u901a\u8fc7\u8de8\u6a21\u6001\u91cd\u6784\u4efb\u52a1\u6765\u8bc4\u4f30\u8fd9\u79cd\u80fd\u529b\uff0c\u6bd4\u5982\u6839\u636e\u5148\u524d\u672a\u89c1\u8fc7\u7684\u7ec6\u80de\u5f62\u6001\u6765\u9884\u6d4b\u57fa\u56e0\u8868\u8fbe\uff0c\u6216\u662f\u9884\u6d4b\u7ec6\u80de\u72b6\u6001\u663e\u5fae\u56fe\u50cf\u5e8f\u5217\u4e2d\u7684\u4e0b\u4e00\u5f20\u56fe\u50cf\u3002\u8bc4\u4f30\u6cdb\u5316\u80fd\u529b\u6709\u52a9\u4e8e\u5efa\u7acb\u4eba\u4eec\u5bf9 AIVC \u6a21\u578b\u7684\u4fe1\u5fc3\uff0c\u76f8\u4fe1\u5b83\u80fd\u591f\u6355\u6349\u6838\u5fc3\u751f\u7269\u5b66\u8fc7\u7a0b\uff0c\u5e76\u7406\u89e3\u8fd9\u4e9b\u8fc7\u7a0b\u5728\u4e0d\u540c\u60c5\u5883\u4e0b\u7684\u53d8\u5316\u3002\u5efa\u7acb\u8fde\u63a5\u751f\u7269\u5b66\u5c3a\u5ea6\u4e0e\u6a21\u6001\u7684\u8de8\u6a21\u6001\u57fa\u51c6\uff0c\u5bf9\u79d1\u7814\u754c\u800c\u8a00\u662f\u4e00\u9879\u5f53\u52a1\u4e4b\u6025\uff0c\u56e0\u4e3a\u8fd9\u4e9b\u4efb\u52a1\u4e0d\u4ec5\u5728\u751f\u7269\u5b66\u4e0a\u5177\u6709\u5b9e\u7528\u4ef7\u503c\uff0c\u800c\u4e14\u5b9a\u4e49\u660e\u786e\u3002<\/span><\/span><\/p>\n

                                            \u6700\u7ec8\uff0c\u8bc4\u5224 AIVC \u6a21\u578b\u5e94\u57fa\u4e8e\u5176<\/span><\/span>\u80fd\u5426\u5f00\u8f9f\u7406\u89e3\u751f\u7269\u5b66\u7684\u65b0\u9014\u5f84<\/span><\/span><\/strong>\u3002\u8fd9\u6837\u7684\u8bc4\u4f30\u6807\u51c6\u80fd\u591f\u786e\u4fdd\u6a21\u578b\u5f00\u53d1\u4e0e\u751f\u7269\u5b66\u76f8\u5173\u6027\u4fdd\u6301\u4e00\u81f4\u3002\u6700\u521d\u6700\u6709\u4ef7\u503c\u7684\u6210\u679c\uff0c\u53ef\u80fd\u5728\u4e8e\u751f\u6210\u6709\u610f\u4e49\u4e14\u53ef\u9a8c\u8bc1\u7684\u5b9e\u9a8c\u5047\u8bbe\u3002\u4e3a\u6b64\uff0c\u4e0e\u5b9e\u9a8c\u53ef\u9a8c\u8bc1\u8868\u578b\u76f8\u5173\u7684\u9a8c\u8bc1\u6570\u636e\u96c6\u53ef\u80fd\u5f88\u9002\u7528\uff0c\u4f8b\u5982\u7ec6\u80de\u7684\u751f\u957f\u901f\u7387\u3001\u5206\u5b50\u8c31\u3001\u88ab\u7834\u574f\u7684\u86cb\u767d\u8d28-\u86cb\u767d\u8d28\u76f8\u4e92\u4f5c\u7528\uff0c\u6216\u8f6c\u5f55\u56e0\u5b50\u7ed3\u5408\u3002<\/span><\/span><\/span><\/p>\n

                                            \u968f\u7740 AIVC \u80fd\u529b\u7684\u63d0\u5347\uff0c\u6211\u4eec\u5fc5\u987b\u8003\u8651\u5355\u7eaf\u7684\u7edf\u8ba1\u6027\u80fd\u6307\u6807\u662f\u5426\u8db3\u591f\uff0c\u6216\u8005<\/span><\/span>\u53ef\u89e3\u91ca\u6027 (interpretability)<\/span><\/span><\/strong>\u548c<\/span><\/span>\u751f\u7269\u5b66\u56e0\u679c\u5173\u7cfb (biological causality)<\/span><\/span><\/strong>\u662f\u5426\u4f1a\u6210\u4e3a\u5176\u6838\u5fc3\u8981\u6c42\u3002<\/span><\/span><\/span><\/p>\n

                                              \n
                                            • \n

                                              \u53ef\u89e3\u91ca\u6027\u548c\u76f8\u4e92\u4f5c\u7528<\/span><\/span><\/strong><\/p>\n<\/li>\n<\/ul>\n

                                              \u5728\u751f\u7269\u5b66\u4e2d\uff0c\u79d1\u5b66\u53d1\u73b0\u7684\u6807\u5fd7\u4e4b\u4e00\u662f\u4e3a\u89c2\u5bdf\u5230\u7684\u73b0\u8c61\u521b\u5efa<\/span><\/span>\u673a\u5236\u6a21\u578b (mechanistic models) <\/span><\/span><\/strong>\u3002\u5728\u521b\u5efa\u865a\u62df\u7ec6\u80de\u65f6\uff0c\u6211\u4eec\u53ef\u80fd\u4e0d\u5f97\u4e0d\u653e\u5f03\u6784\u5efa\u5b8c\u5168\u673a\u5236\u6a21\u578b\u7684\u80fd\u529b\uff0c\u8f6c\u800c\u5b66\u4e60\u80fd\u591f\u4ece\u6570\u636e\u4e2d\u6cdb\u5316\u5e76\u9884\u6d4b\u8d85\u51fa\u89c2\u5bdf\u8303\u56f4\u7684\u76f8\u4e92\u4f5c\u7528\u3002\u7136\u800c\uff0c\u52aa\u529b\u63d0\u9ad8\u53ef\u89e3\u91ca\u6027\u4ecd\u7136\u662f\u53ef\u53d6\u7684\u3002<\/span><\/span><\/span><\/p>\n

                                              \u6bcf\u4e2a AIVC \u9884\u6d4b\u90fd\u53ef\u4ee5\u901a\u8fc7\u51b3\u5b9a\u7ed3\u679c\u72b6\u6001\u7684\u76f8\u5e94\u591a\u5c3a\u5ea6\u76f8\u4e92\u4f5c\u7528\u6765\u8bc1\u5b9e\u3002\u4f8b\u5982\uff0c\u4e86\u89e3\u7ec6\u80de\u4e9a\u7cfb\u7edf\u6216\u86cb\u767d\u8d28\u590d\u5408\u7269\u5728\u75be\u75c5\u7ec4\u7ec7\u4e2d\u5982\u4f55\u88ab\u7834\u574f\uff0c\u53ef\u4ee5\u5e2e\u52a9\u5f00\u53d1\u6cbb\u7597\u5e72\u9884\u63aa\u65bd [118,119]\u3002AIVC \u7684\u6a21\u5757\u5316\u7ed3\u6784\u5c06\u4f7f\u7814\u7a76\u4eba\u5458\u80fd\u591f\u7cbe\u786e\u5b9a\u4f4d\u53c2\u4e0e\u6bcf\u4e2a\u9884\u6d4b\u884c\u4e3a\u7684\u7279\u5b9a\u57fa\u56e0\u3001\u86cb\u767d\u8d28\u6216\u5206\u5b50\u8fc7\u7a0b\u3002<\/span><\/span><\/p>\n

                                              \u5927\u578b\u6a21\u578b\u5e03\u7ebf\u4e2d\u7684\u6a21\u5f0f\u4e5f\u53ef\u4ee5\u88ab\u5229\u7528\u6765\u63ed\u793a\u7ec4\u5408\u751f\u7269\u5b66\u76f8\u4e92\u4f5c\u7528\uff0c\u4f8b\u5982\u86cb\u767d\u8d28\u4e4b\u95f4\u7684\u76f8\u4e92\u4f5c\u7528\uff0c\u8fd9\u4e9b\u76f8\u4e92\u4f5c\u7528\u53ef\u4ee5\u6295\u5c04\u5230\u53ef\u89e3\u91ca\u7684\u7a7a\u95f4\u4e2d\uff0c\u800c\u4e0d\u9650\u5236\u539f\u59cb\u6a21\u578b\u7684\u901a\u7528\u6027\u3002\u5c3d\u7ba1 AIVC \u7684\u8bb8\u591a\u80fd\u529b\u4f9d\u8d56\u4e8e\u9884\u6d4b\u4efb\u52a1\uff0c\u4f46\u751f\u6210\u673a\u5236\u5047\u8bbe\u53ef\u4ee5\u4e3a\u7406\u89e3\u548c\u8fdb\u4e00\u6b65\u63a2\u7d22 AIVC \u7684\u9884\u6d4b\u63d0\u4f9b\u5b9e\u9a8c\u9014\u5f84\uff0c\u8fd9\u5bf9\u4e8e AIVC \u7684\u91c7\u7528\u548c\u4f7f\u7528\u81f3\u5173\u91cd\u8981\u3002<\/span><\/span><\/p>\n

                                              \u6700\u7ec8\uff0c\u4e3a AIVC \u6784\u5efa\u4e00\u4e2a\u4ea4\u4e92\u5c42<\/span> (interactive layer) <\/span>\u5c06\u662f\u5173\u952e\u6240\u5728\uff0c\u8fd9\u5c06\u4f7f\u4e0d\u540c\u4e13\u4e1a\u6c34\u5e73\u7684\u7814\u7a76\u4eba\u5458\u80fd\u591f\u6709\u6548\u5730\u7406\u89e3\u548c\u5229\u7528\u5176\u9884\u6d4b\u7ed3\u679c\u3002\u4f7f\u7528 LLMs \u6784\u5efa\u7684<\/span><\/span>AI\u667a\u80fd\u4f53 (AI agents)<\/span><\/span><\/strong> \u53ef\u4ee5\u4f5c\u4e3a\u865a\u62df\u7814\u7a76\u52a9\u624b\uff0c\u4e3a\u975e\u4e13\u4e1a\u4eba\u5458\u63d0\u4f9b\u76f4\u89c2\u7684\u754c\u9762 [46,120]\u3002\u5229\u7528\u5b83\u4eec\u5bf9\u79d1\u5b66\u6587\u732e\u7684\u5e7f\u6cdb\u77e5\u8bc6\uff0c\u8fd9\u4e9b\u8bed\u8a00\u6a21\u578b\u53ef\u4ee5\u4e3a AIVC \u505a\u51fa\u7684\u9884\u6d4b\u63d0\u4f9b\u66f4\u6df1\u5165\u7684\u89c1\u89e3\u3002<\/span><\/span><\/span><\/p>\n


                                              <\/span><\/h3>\n

                                              \n
                                              \n
                                              \n
                                              \n
                                              \n

                                              <\/span><\/section>\n

                                              <\/span><\/section>\n
                                              \n
                                              <\/section>\n<\/section>\n<\/section>\n
                                              \n

                                              7. \u524d\u666f\u5c55\u671b<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

                                              \n

                                              <\/span><\/section>\n

                                              <\/span><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                                              <\/span><\/p>\n

                                              \u9057\u4f20\u5b66\u548c\u57fa\u56e0\u7ec4\u5b66\u754c\u5df2\u7ecf\u521b\u5efa\u4e86\u5927\u578b\u53c2\u8003\u6570\u636e\u96c6\uff0c\u5982\u4eba\u7c7b\u57fa\u56e0\u7ec4\u8ba1\u5212<\/span> (human genome project) <\/span>[23]\u3001HapMap [122]\u3001\u764c\u75c7\u57fa\u56e0\u7ec4\u56fe\u8c31<\/span> (the Cancer Genome Atlas, TCGA) <\/span> [123]\u3001ENCODE [124]\u3001\u57fa\u56e0\u578b-\u7ec4\u7ec7\u8868\u8fbe <\/span> (the Genotype-Tissue Expression, GTEx) <\/span>\u9879\u76ee [125]\u3001\u4eba\u7c7b\u86cb\u767d\u8d28\u56fe\u8c31<\/span> (the Human Protein Atlas, HPA) <\/span>[64,126]\u3001\u4eba\u7c7b\u7ec6\u80de\u56fe\u8c31<\/span> (the Human Cell Atlas, HCA) <\/span> [24]\uff0c\u4ee5\u53ca\u8d8a\u6765\u8d8a\u591a\u7684\u6df1\u5ea6\u8868\u578b\u5316\u3001\u4eba\u7fa4\u89c4\u6a21\u7684\u751f\u7269\u6837\u672c\u5e93\u5de5\u4f5c [127]\u3002\u5f97\u76ca\u4e8e\u8fd9\u4e9b\u9879\u76ee\uff0c\u73b0\u5728\u6709\u5927\u91cf\u53c2\u8003\u6570\u636e\u53ef\u7528\u4e8e\u8bad\u7ec3\u673a\u5668\u5b66\u4e60\u6a21\u578b\u3002\u5c3d\u7ba1\u8fd9\u4e9b\u52aa\u529b\u5c06\u7ee7\u7eed\u53d1\u5c55\uff0c\u4f46\u5b83\u4eec\u4e5f\u50ac\u5316\u4e86\u4e00\u9879\u65b0\u7684\u5e76\u884c\u5de5\u4f5c\uff1a\u521b\u5efa\u7ec6\u80de\u751f\u7269\u5b66\u7684\u865a\u62df\u4eff\u771f\uff0c\u8fd9\u662f\u4e00\u79cd\u65b0\u7684\u79d1\u5b66\u63a2\u7a76\u8fc7\u7a0b\u3002<\/span><\/span><\/p>\n

                                              \u56e0\u6b64\uff0cAIVC \u6709\u6f5c\u529b\u5f7b\u5e95\u6539\u53d8\u79d1\u5b66\u7814\u7a76\u8fc7\u7a0b\uff0c\u5728\u751f\u7269\u533b\u5b66\u7814\u7a76\u3001\u4e2a\u6027\u5316\u533b\u7597\u3001\u836f\u7269\u53d1\u73b0\u3001\u7ec6\u80de\u5de5\u7a0b\u548c\u53ef\u7f16\u7a0b\u751f\u7269\u5b66<\/span> (programmable biology) <\/span>\u9886\u57df\u5e26\u6765\u672a\u6765\u7684\u7a81\u7834\u3002\u4f5c\u4e3a\u865a\u62df\u5b9e\u9a8c\u5ba4<\/span> (virtual laboratory)<\/span>\uff0cAIVC \u53ef\u4ee5\u4fc3\u8fdb\u8ba1\u7b97\u673a\u6a21\u62df\u5b9e\u9a8c\u6570\u636e\u4e0e\u7269\u7406\u5b9e\u9a8c\u5ba4\u7ed3\u679c\u4e4b\u95f4\u7684\u65e0\u7f1d\u5bf9\u63a5\u3002\u56e0\u6b64\uff0c\u6211\u4eec\u671f\u671b AIVC \u80fd\u591f\u4e3a\u751f\u7269\u8fc7\u7a0b\u63d0\u4f9b\u66f4\u7edf\u4e00\u7684\u89c6\u89d2\uff0c\u4fc3\u8fdb\u79d1\u5b66\u5bb6\u4eec\u5728\u751f\u7269\u5b66\u6d8c\u73b0\u7279\u6027\u5982\u4f55\u4ea7\u751f\u8fd9\u4e00\u95ee\u9898\u4e0a\u8fbe\u6210\u5171\u8bc6\u3002<\/span><\/span><\/p>\n

                                              \u901a\u8fc7\u8fde\u63a5\u8ba1\u7b97\u673a\u7cfb\u7edf\u3001\u73b0\u4ee3\u751f\u6210\u5f0f AI \u548c AI \u667a\u80fd\u4f53\u4ee5\u53ca\u751f\u7269\u5b66\u7684\u4e16\u754c\uff0cAIVC \u6700\u7ec8\u80fd\u591f\u8ba9\u79d1\u5b66\u5bb6\u5c06\u7ec6\u80de\u7406\u89e3\u4e3a\u4fe1\u606f\u5904\u7406\u7cfb\u7edf\uff0c\u5e76\u6784\u5efa\u751f\u547d\u7684\u865a\u62df\u63cf\u8ff0\u3002\u968f\u7740 AIVC \u6269\u5c55\u5bf9\u7ec6\u80de\u548c\u5206\u5b50\u7cfb\u7edf\u7684\u7406\u89e3\uff0c\u5b83\u4e5f\u5c06\u8d8a\u6765\u8d8a\u591a\u5730\u5141\u8bb8\u6211\u4eec\u5bf9\u5b83\u4eec\u8fdb\u884c\u7f16\u7a0b\u5e76\u8bbe\u8ba1\u65b0\u9896\u7684\u5408\u6210\u7cfb\u7edf\u3002AI \u6a21\u578b\u5df2\u7ecf\u88ab\u7528\u4e8e\u8bbe\u8ba1\u65b0\u7684 CRISPR \u9176 [53]\u3001\u529f\u80fd\u86cb\u767d [128]\uff0c\u751a\u81f3\u6574\u4e2a\u539f\u6838\u751f\u7269\u57fa\u56e0\u7ec4 [51]\u3002\u7ec6\u80de\u548c\u57fa\u56e0\u7ec4\u5de5\u7a0b\u5de5\u5177\u7cbe\u5ea6\u7684\u5feb\u901f\u8fdb\u6b65\u5c06\u52a0\u901f\u8fd9\u4e00\u8f6c\u53d8\uff0cAIVC \u7684\u4e0d\u540c\u5b9e\u4f8b\u5316\u5c06\u5728\u5de5\u7a0b\u65b0\u7684\u529f\u80fd\u6027\u751f\u7269\u5b66\u80fd\u529b\u65b9\u9762\u5c55\u5f00\u7ade\u4e89\uff0c\u5c31\u50cf\u5728\u8868\u793a\u548c\u6a21\u62df\u751f\u7269\u5b66\u65b9\u9762\u7684\u7ade\u4e89\u4e00\u6837\u6fc0\u70c8\u3002<\/span><\/span><\/p>\n


                                              <\/span><\/p>\n

                                              \n
                                              \n
                                              \n

                                              \u53c2\u8003\u6587\u732e<\/span><\/strong><\/p>\n<\/section>\n<\/section>\n<\/section>\n

                                              \n
                                              \n
                                              \n
                                              \n


                                              <\/span><\/p>\n

                                              \n
                                              \n
                                                \n
                                              1. \n

                                                Slepchenko, B.M., Schaff, J.C., Macara, I., and Loew, L.M. (2003). Quan-<\/span><\/span><\/span><\/span>titative cell biology with the Virtual Cell. Trends Cell Biol. 13, 570\u2013576<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              2. \n

                                                Johnson, G.T., Agmon, E., Akamatsu, M., Lundberg, E., Lyons, B.,<\/span><\/span><\/span><\/span>Ouyang, W., Quintero-Carmona, O.A., Riel-Mehan, M., Rafelski, S.,<\/span><\/span><\/span><\/span>and Horwitz, R. (2023). Building the next generation of virtual cells to un-<\/span><\/span><\/span><\/span>derstand cellular biology. Biophys. J. 122, 3560\u20133569<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              3. \n

                                                Marx, V. (2023). How to build a virtual embryo. Nat. Methods 20,<\/span><\/span><\/span><\/span>1838\u20131843<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              4. \n

                                                Goldberg, A.P., Szigeti, B., Chew, Y.H., Sekar, J.A., Roth, Y.D., and Karr,J.R. (2018). Emerging whole-cell modeling principles and methods. Curr.<\/span><\/span><\/span><\/span>Opin. Biotechnol. 51, 97\u2013102<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              5. \n

                                                Georgouli, K., Yeom, J.-S., Blake, R.C., and Navid, A. (2023). Multi-scale<\/span><\/span><\/span><\/span>models of whole cells: progress and challenges. Front. Cell Dev. Biol. 11,<\/span><\/span><\/span><\/span>1260507<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              6. \n

                                                Marucci, L., Barberis, M., Karr, J., Ray, O., Race, P.R., de Souza Andrade,<\/span><\/span><\/span><\/span>M., Grierson, C., Hoffmann, S.A., Landon, S., Rech, E., et al. (2020). Com-<\/span><\/span><\/span><\/span>puter-aided whole-cell design: Taking a holistic approach by integrating<\/span><\/span><\/span><\/span>synthetic with systems biology. Front. Bioeng. Biotechnol. 8, 942<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              7. \n

                                                Lauffenburger, D.A., and Linderman, J.J. (1996). Receptors: models for<\/span><\/span><\/span><\/span>binding, trafficking, and signaling (Oxford University Press)<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              8. \n

                                                Karr, J.R., Sanghvi, J.C., Macklin, D.N., Gutschow, M.V., Jacobs, J.M.,<\/span><\/span><\/span><\/span>Bolival, B., Assad-Garcia, N., Glass, J.I., and Covert, M.W. (2012). A<\/span><\/span><\/span><\/span>whole-cell computational model predicts phenotype from genotype.<\/span><\/span><\/span><\/span>Cell 150, 389\u2013401<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              9. \n

                                                Mangan, S., and Alon, U. (2003). Structure and function of the feed-for-<\/span><\/span><\/span><\/span>ward loop network motif. Proc. Natl. Acad. Sci. USA 100, 11980\u201311985<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              10. \n

                                                Zopf, C.J., Quinn, K., Zeidman, J., and Maheshri, N. (2013). Cell-cycle<\/span><\/span><\/span><\/span>dependence of transcription dominates noise in gene expression.<\/span><\/span><\/span><\/span>PLoS Comput. Biol. 9, e1003161<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              11. \n

                                                Eling, N., Morgan, M.D., and Marioni, J.C. (2019). Challenges in<\/span><\/span><\/span><\/span>measuring and understanding biological noise. Nat. Rev. Genet. 20,<\/span><\/span><\/span><\/span>536\u2013548<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              12. \n

                                                Hellweger, F.L., Clegg, R.J., Clark, J.R., Plugge, C.M., and Kreft, J.-U.<\/span><\/span><\/span><\/span>(2016). Advancing microbial sciences by individual-based modelling.<\/span><\/span><\/span><\/span>Nat. Rev. Microbiol. 14, 461\u2013471<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              13. \n

                                                Gorochowski, T.E. (2016). Agent-based modelling in synthetic biology.<\/span><\/span><\/span><\/span>Essays Biochem. 60, 325\u2013336<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              14. \n

                                                Thiele, I., Jamshidi, N., Fleming, R.M., and Palsson, B. (2009). Genome-<\/span><\/span><\/span><\/span>scale reconstruction of Escherichia coli\u2019s transcriptional and translational<\/span><\/span><\/span><\/span>machinery: a knowledge base, its mathematical formulation, and its func-<\/span><\/span><\/span><\/span>tional characterization. PLoS Comput. Biol. 5, e1000312<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              15. \n

                                                Odell, G.M., and Foe, V.E. (2008). An agent-based model contrasts<\/span><\/span><\/span><\/span>opposite effects of dynamic and stable microtubules on cleavage furrow<\/span><\/span><\/span><\/span>positioning. J. Cell Biol. 183, 471\u2013483<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              16. \n

                                                Popov, K., Komianos, J., and Papoian, G.A. (2016). MEDYAN: mechano-<\/span><\/span><\/span><\/span>chemical simulations of contraction and polarity alignment in actomyosin<\/span><\/span><\/span><\/span>networks. PLoS Comput. Biol. 12, e1004877<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              17. \n

                                                Burke, P.E.P., Campos, C.B.L., Costa, L.D.F., and Quiles, M.G. (2020).M. G. A biochemical network modeling of a whole-cell. Sci. Rep.10, 13303<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              18. \n

                                                Li, G., Liu, L., Du, W., and Cao, H. (2023). Local flux coordination and<\/span><\/span><\/span><\/span>global gene expression regulation in metabolic modeling. Nat. Commun.<\/span><\/span><\/span><\/span>14, 5700<\/span><\/span><\/em><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              19. \n

                                                Fang, X., Lloyd, C.J., and Palsson, B.O. (2020). Reconstructing organ-<\/span><\/span><\/span><\/span>isms in silico: genome-scale models and their emerging applications.<\/span><\/span><\/span><\/span>Nat. Rev. Microbiol. 18, 731\u2013743<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              20. \n

                                                Stevens, J.A., Gru\u00a8 newald, F., van Tilburg, P.A.M., Ko\u00a8 nig, M., Gilbert,<\/span><\/span><\/span><\/span>B.R., Brier, T.A., Thornburg, Z.R., Luthey-Schulten, Z., and Marrink,S.J. (2023). Molecular dynamics simulation of an entire cell. Front.<\/span><\/span><\/span><\/span>Chem. 11, 1106495<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              21. \n

                                                Maritan, M., Autin, L., Karr, J., Covert, M.W., Olson, A.J., and Goodsell,D.S. (2022). Building structural models of a whole mycoplasma cell.J. Mol. Biol. 434, 167351<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              22. \n

                                                Ahn-Horst, T.A., Mille, L.S., Sun, G., Morrison, J.H., and Covert, M.W.<\/span><\/span><\/span><\/span>(2022). An expanded whole-cell model of E. coli links cellular physiology<\/span><\/span><\/span><\/span>with mechanisms of growth rate control. npj Syst. Biol. Appl. 8, 30<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              23. \n

                                                Venter, J.C., Adams, M.D., Myers, E.W., Li, P.W., Mural, R.J., Sutton,G.G., Smith, H.O., Yandell, M., Evans, C.A., Holt, R.A., et al. (2001).<\/span><\/span><\/span><\/span>The sequence of the human genome. Science 291, 1304\u20131351<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              24. \n

                                                Regev, A., Teichmann, S.A., Lander, E.S., Amit, I., Benoist, C., Birney, E.,<\/span><\/span><\/span><\/span>Bodenmiller, B., Campbell, P., Carninci, P., Clatworthy, M., et al. (2017).<\/span><\/span><\/span><\/span>The human cell atlas. eLife 6, e27041<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              25. \n

                                                CZI Single-Cell Biology Program, Abdulla, S., Aevermann, B., Assis, P.,<\/span><\/span><\/span><\/span>Badajoz, S., Bell, S.M., Bezzi, E., Batuhan, C., Jim, C., Chambers, S.,<\/span><\/span><\/span><\/span>et al. (2023). CZ CELL3GENE discover: A single- cell data platform for<\/span><\/span><\/span><\/span>scalable exploration, analysis and modeling of aggregated data. Preprint<\/span><\/span><\/span><\/span>at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              26. \n

                                                Heimberg, G., Kuo, T., DePianto, D., Heigl, T., Nathaniel, D., Salem, O.,<\/span><\/span><\/span><\/span>Scalia, G., Biancalani, T., Turley, S., Rock, J., et al. (2023). Scalable<\/span><\/span><\/span><\/span>querying of human cell atlases via a foundational model reveals com-<\/span><\/span><\/span><\/span>monalities across fibrosis-associated macrophages. Preprint at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              27. \n

                                                Dixit, A., Parnas, O., Li, B., Chen, J., Fulco, C.P., Jerby-Arnon, L., Marja-<\/span><\/span><\/span><\/span>novic, N.D., Dionne, D., Burks, T., Raychowdhury, R., et al. (2016). Per-<\/span><\/span><\/span><\/span>turb-Seq: dissecting molecular circuits with scalable single-cell RNA<\/span><\/span><\/span><\/span>profiling of pooled genetic screens. Cell 167, 1853\u20131866.e17<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              28. \n

                                                Srivatsan, S.R., McFaline-Figueroa, J.L., Ramani, V., Saunders, L., Cao,<\/span><\/span><\/span><\/span>J., Packer, J., Pliner, H.A., Jackson, D.L., Daza, R.M., Christiansen, L.,<\/span><\/span><\/span><\/span>et al. (2020). Massively multiplex chemical transcriptomics at single-<\/span><\/span><\/span><\/span>cell resolution. Science 367, 45\u201351<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              29. \n

                                                Feldman, D., Funk, L., Le, A., Carlson, R.J., Leiken, M.D., Tsai, F., Soong,<\/span><\/span><\/span><\/span>B., Singh, A., and Blainey, P.C. (2022). Pooled genetic perturbation<\/span><\/span><\/span><\/span>screens with image-based phenotypes. Nat. Protoc. 17, 476\u2013512<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              30. \n

                                                Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez,<\/span><\/span><\/span><\/span>A.N., Kaiser, L., and Polosukhin, I. (2017). Attention is all you need.<\/span><\/span><\/span><\/span>Adv. Neural Inf. Process. Syst. 30<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              31. \n

                                                Rombach, R., Blattmann, A., Lorenz, D., Esser, P., and Ommer, B. High- resolution image synthesis with latent diffusion models. In IEEE Confer- ence on Computer Vision and Pattern Recognition (CVPR), 10684\u201310695.<\/span><\/span><\/p>\n<\/li>\n

                                              32. \n

                                                Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger,<\/span><\/span><\/span><\/span>O., Tunyasuvunakool, K., Bates, R.,<\/span><\/span><\/span> Z\u02c7<\/span><\/span><\/span>\u00b4\u0131dek, A., Potapenko, A., et al.<\/span><\/span><\/span><\/span><\/span>(2021). Highly accurate protein structure prediction with AlphaFold. Na-<\/span><\/span><\/span><\/span>ture 596, 583\u2013589<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              33. \n

                                                Baek, M., DiMaio, F., Anishchenko, I., Dauparas, J., Ovchinnikov, S., Lee,<\/span><\/span><\/span><\/span>G.R., Wang, J., Cong, Q., Kinch, L.N., Schaeffer, R.D., et al. (2021). Ac-<\/span><\/span><\/span><\/span>curate prediction of protein structures and interactions using a three-<\/span><\/span><\/span><\/span>track neural network. Science 373, 871\u2013876<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              34. \n

                                                Lin, Z., Akin, H., Rao, R., Hie, B., Zhu, Z., Lu, W., Smetanin, N., Verkuil, R.,<\/span><\/span><\/span><\/span>Kabeli, O., Shmueli, Y., et al. (2023). Evolutionary-scale prediction of<\/span><\/span><\/span><\/span>atomic-level protein structure with a language model. Science 379,<\/span><\/span><\/span><\/span>1123\u20131130<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              35. \n

                                                Gomes, J., Ramsundar, B., Feinberg, E.N., and Pande, V.S. (2017).<\/span><\/span><\/span><\/span>Atomic convolutional networks for predicting protein-ligand binding af-<\/span><\/span><\/span><\/span>finity. Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              36. \n

                                                Cunningham, J.M., Koytiger, G., Sorger, P.K., and AlQuraishi, M. (2020).<\/span><\/span><\/span><\/span>Biophysical prediction of protein\u2013peptide interactions and signaling net-<\/span><\/span><\/span><\/span>works using machine learning. Nat. Methods 17, 175\u2013183<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              37. \n

                                                Torng, W., and Altman, R.B. (2019). High precision protein functional site<\/span><\/span><\/span><\/span>detection using 3D convolutional neural networks. Bioinformatics 35,<\/span><\/span><\/span><\/span>1503\u20131512<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              38. \n

                                                Corso G., Sta\u00a8 rk H., Jing B., Barzilay R., and Jaakkola T. (2023). DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking The Eleventh International Conference on Learning Representations.<\/span><\/span><\/p>\n<\/li>\n

                                              39. \n

                                                Kudo, T., Meireles, A.M., Moncada, R., Chen, Y., Wu, P., Gould, J., Hu,<\/span><\/span><\/span><\/span>X., Kornfeld, O., Jesudason, R., Foo, C., et al. (2024). Multiplexed, im-<\/span><\/span><\/span><\/span>age-based pooled screens in primary cells and tissues with perturbview.<\/span><\/span><\/span><\/span>Nat. Biotechnol., 1\u201310<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              40. \n

                                                Roohani, Y., Huang, K., and Leskovec, J. (2023). Predicting transcrip-<\/span><\/span><\/span><\/span>tional outcomes of novel multigene perturbations with GEARS. Nat. Bio-<\/span><\/span><\/span><\/span>technol. 42, 927\u2013935<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              41. \n

                                                Bunne, C., Stark, S.G., Gut, G., Del Castillo, J.S., Levesque, M., Leh-<\/span><\/span><\/span><\/span>mann, K.-V., Pelkmans, L., Krause, A., and Ratsch, G. (2023). Learning<\/span><\/span><\/span><\/span>single-cell perturbation responses using neural optimal transport. Nat.<\/span><\/span><\/span><\/span>Methods 20, 1759\u20131768<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              42. \n

                                                Lotfollahi, M., Klimovskaia Susmelj, A., De Donno, C., Hetzel, L., Ji, Y.,<\/span><\/span><\/span><\/span>Ibarra, I.L., Srivatsan, S.R., Naghipourfar, M., Daza, R.M., Martin, B.,<\/span><\/span><\/span><\/span>et al. (2023). Predicting cellular responses to complex perturbations in<\/span><\/span><\/span><\/span>high-throughput screens. Mol. Syst. Biol. 19, e11517<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              43. \n

                                                Bunne, C., Krause, A., and Cuturi, M. (2022). Supervised Training of Con-<\/span><\/span><\/span><\/span>ditional Monge Maps. Adv. Neural Inf. Process. Syst. 35, 6859\u20136872<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              44. \n

                                                Bereket, M., and Karaletsos, T. (2024). Modelling Cellular Perturbations<\/span><\/span><\/span><\/span>with the Sparse Additive Mechanism Shift Variational Autoencoder.<\/span><\/span><\/span><\/span>Adv. Neural Inf. Process. Syst. 36<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              45. \n

                                                Huang, K., Huang, K., Lopez, R., Hutter, J.-C., Kudo, T., Rios, A., and Re-<\/span><\/span><\/span><\/span>gev, A. (2023). Sequential Optimal Experimental De- sign of Perturbation<\/span><\/span><\/span><\/span>Screens Guided by Multi-modal Priors. In International Conference on<\/span><\/span><\/span><\/span>Research in Computational Molecular Biology (Springer Nature),<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              46. \n

                                                Roohani, Y.H., Vora, J., Huang, Q., Liang, P., and Leskovec, J. (2024).<\/span><\/span><\/span><\/span>BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Ex-<\/span><\/span><\/span><\/span>periments Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              47. \n

                                                Cleary, B., and Regev, A. (2024). The necessity and power of random,<\/span><\/span><\/span><\/span>undersampled experiments in biology. Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              48. \n

                                                Ji, Y., Zhou, Z., Liu, H., and Davuluri, R.V. (2021). DNABERT: pre-trained<\/span><\/span><\/span><\/span>Bidirectional Encoder Representations from Transformers model for<\/span><\/span><\/span><\/span>DNA-language in genome. Bioinformatics 37, 2112\u20132120<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              49. \n

                                                Brandes, N., Ofer, D., Peleg, Y., Rappoport, N., and Linial, M. (2022). Pro-<\/span><\/span><\/span><\/span>teinBERT: a universal deep-learning model of pro- tein sequence and<\/span><\/span><\/span><\/span>function. Bioinformatics 38, 2102\u20132110<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              50. \n

                                                Celaj, A., Gao, A.J., Lau, T.T.Y., Holgersen, E.M., Lo, A., Lodaya, V., Cole,<\/span><\/span><\/span><\/span>C.B., Denroche, R.E., Spickett, C., Wagih, O., et al. (2023). An RNA foun-<\/span><\/span><\/span><\/span>dation model enables discovery of disease mechanisms and candidate<\/span><\/span><\/span><\/span>therapeutics. Preprint at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              51. \n

                                                Nguyen, E., Poli, M., Durrant, M.G., Kang, B., Katrekar, D., Li, D.B., Bar-tie, L.J., Thomas, A.W., King, S.H., Brixi, G., et al. (2024). Sequence<\/span><\/span><\/span><\/span>modeling and design from molecular to genome scale with Evo. Science<\/span><\/span><\/span><\/span>386, eado9336<\/span><\/span><\/em><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              52. \n

                                                Hayes, T., Rao, R., Akin, H., Sofroniew, N.J., Oktay, D., Lin, Z., Verkuil, R.,<\/span><\/span><\/span><\/span>Tran, V.Q., Deaton, J., Wiggert, M., et al. (2024). Simulating 500 million<\/span><\/span><\/span><\/span>years of evolution with a language model. Preprint at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              53. \n

                                                Ruffolo, J.A., Nayfach, S., Gallagher, J., Bhatnagar, A., Beazer, J., Hus-<\/span><\/span><\/span><\/span>sain, R., Russ, J., Yip, J., Hill, E., Pacesa, M., et al. (2024). Design of highly<\/span><\/span><\/span><\/span>functional genome editors by modeling the universe of CRISPR-cas se-<\/span><\/span><\/span><\/span>quences. Preprint at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              54. \n

                                                Peng, Z., Schussheim, B., and Chatterjee, P. (2024). PTM-mamba: a<\/span><\/span><\/span><\/span>PTM-aware protein language model with bidirectional gated mamba<\/span><\/span><\/span><\/span>blocks. Preprint at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              55. \n

                                                Dai, B., Mattox, D.E., and Bailey-Kellogg, C. (2021). Atten- tion please:<\/span><\/span><\/span><\/span>modeling global and local context in glycan structure-function relation-<\/span><\/span><\/span><\/span>ships. Preprint at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              56. \n

                                                Yu, T., Yao, T., Sun, Z., Shi, F., Zhang, L., Lyu, K., Xuan, B., Liu, A., Zhang,<\/span><\/span><\/span><\/span>X., Zou, J., et al. (2024). LipidBERT: A Lipid Language Model Pre- trained<\/span><\/span><\/span><\/span>on METiS de novo Lipid Library. Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              57. \n

                                                Krishna, R., Wang, J., Ahern, W., Sturmfels, P., Venkatesh, P., Kalvet, I.,<\/span><\/span><\/span><\/span>Lee, G.R., Morey-Burrows, F.S., Anishchenko, I., Humphreys, I.R., et al.<\/span><\/span><\/span><\/span>(2024). Generalized biomolecular modeling and design with<\/span><\/span><\/span><\/span>RoseTTAFold All-Atom. Science 384, eadl2528<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              58. \n

                                                Rosen, Y., Brbi<\/span><\/span><\/span>c\u00b4<\/span><\/span><\/span>, M., Roohani, Y., Swanson, K., Li, Z., and Leskovec, J.<\/span><\/span><\/span><\/span><\/span>(2024). Toward universal cell embeddings: integrating single-cell RNA-<\/span><\/span><\/span><\/span>seq datasets across species with Saturn. Nat. Methods 21, 1492\u20131500<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              59. \n

                                                Rosen, Y., Roohani, Y., Agrawal, A., Samotor<\/span><\/span><\/span>c\u02c7<\/span><\/span><\/span>an, L., Tabula Sapiens<\/span><\/span><\/span><\/span><\/span>Consortium, Quake, S.R., and Leskovec, J. (2023). Universal cell embed-<\/span><\/span><\/span><\/span>dings: A foundation model for cell biology. Preprint at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              60. \n

                                                Chen, Y., and Zou, J. (2024). GenePT: A Simple but Effective Foun- da-<\/span><\/span><\/span><\/span>tion Model for Genes and Cells Built from ChatGPT. Preprint at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              61. \n

                                                Mahdessian, D., Cesnik, A.J., Gnann, C., Danielsson, F., Stenstro\u00a8 m, L.,<\/span><\/span><\/span><\/span>Arif, M., Zhang, C., Le, T., Johansson, F., Schutten, R., et al. (2021).<\/span><\/span><\/span><\/span>Spatiotemporal dissection of the cell cycle with single-cell proteogenom-<\/span><\/span><\/span><\/span>ics. Nature 590, 649\u2013654<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              62. \n

                                                Chandrasekaran, S.N., Cimini, B.A., Goodale, A., Miller, L., Kost-Ali-<\/span><\/span><\/span><\/span>mova, M., Jamali, N., Doench, J.G., Fritchman, B., Skepner, A., Melan-<\/span><\/span><\/span><\/span>son, M., et al. (2024). Three million images and mor- phological profiles<\/span><\/span><\/span><\/span>of cells treated with matched chemical and genetic perturbations. Nat.<\/span><\/span><\/span><\/span>Methods 21, 1114\u20131121<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              63. \n

                                                Carlson, R.J., Leiken, M.D., Guna, A., Hacohen, N., and Blainey, P.C.<\/span><\/span><\/span><\/span>(2023). A genome-wide optical pooled screen reveals regulators of<\/span><\/span><\/span><\/span>cellular antiviral responses. Proc. Natl. Acad. Sci. USA 120,<\/span><\/span><\/span><\/span>e2210623120<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              64. \n

                                                Thul, P.J., A<\/span><\/span><\/span>\u02da <\/span><\/span><\/span>kesson, L., Wiking, M., Mahdessian, D., Geladaki, A., Ait Blal,<\/span><\/span><\/span><\/span><\/span>H., Alm, T., Asplund, A., Bjo\u00a8 rk, L., Breckels, L.M., et al. (2017). A subcel-<\/span><\/span><\/span><\/span>lular map of the human proteome. Science 356, eaal3321<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              65. \n

                                                McDole, K., Guignard, L., Amat, F., Berger, A., Malandain, G., Royer,<\/span><\/span><\/span><\/span>L.A., Turaga, S.C., Branson, K., and Keller, P.J. (2018). In toto imaging<\/span><\/span><\/span><\/span>and reconstruction of post-implantation mouse development at the sin-<\/span><\/span><\/span><\/span>gle-cell level. Cell 175, 859\u2013876.e33<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              66. \n

                                                Nogales, E., and Mahamid, J. (2024). Bridging structural and cell biology<\/span><\/span><\/span><\/span>with cryo-electron microscopy. Nature 628, 47\u201356<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              67. \n

                                                Bauda, E., Gallet, B., Moravcova, J., Effantin, G., Chan, H., Novacek, J.,<\/span><\/span><\/span><\/span>Jouneau, P.H., Rodrigues, C.D.A., Schoehn, G., Moriscot, C., et al.<\/span><\/span><\/span><\/span>(2024). Ultrastructure of macromolecular assemblies contributing to bac-<\/span><\/span><\/span><\/span>terial spore resistance revealed by in situ cryo-electron tomography. Nat.<\/span><\/span><\/span><\/span>Commun. 15, 1376<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              68. \n

                                                Lelek, M., Gyparaki, M.T., Beliu, G., Schueder, F., Griffie\u00b4 , J., Manley, S.,<\/span><\/span><\/span><\/span>Jungmann, R., Sauer, M., Lakadamyali, M., and Zimmer, C. (2021). Sin-<\/span><\/span><\/span><\/span>gle-molecule localization microscopy. Nat. Rev. Methods Primers 1, 39<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              69. \n

                                                Mo\u00a8 ckl, L., and Moerner, W.E. (2020). Super-resolution microscopy with<\/span><\/span><\/span><\/span>single molecules in biology and beyond\u2013essentials, current trends, and<\/span><\/span><\/span><\/span>future challenges. J. Am. Chem. Soc. 142, 17828\u201317844<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              70. \n

                                                Cesnik, A., Schaffer, L.V., Gaur, I., Jain, M., Ideker, T., and Lundberg, E.<\/span><\/span><\/span><\/span>(2024). Mapping the multiscale proteomic Or- ganization of cellular and<\/span><\/span><\/span><\/span>Disease Phenotypes. Annu. Rev. Biomed. Data Sci. 7, 369\u2013389<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              71. \n

                                                Qin, Y., Huttlin, E.L., Winsnes, C.F., Gosztyla, M.L., Wacheul, L., Kelly,<\/span><\/span><\/span><\/span>M.R., Blue, S.M., Zheng, F., Chen, M., Schaffer, L.V., et al. (2021). A<\/span><\/span><\/span><\/span>multi-scale map of cell structure fusing protein images and interactions.<\/span><\/span><\/span><\/span>Nature 600, 536\u2013542<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              72. \n

                                                Dosovitskiy, A. (2020). An image is worth 16×16 words: transformers for<\/span><\/span><\/span><\/span>image recognition at scale. Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              73. \n

                                                Fukushima, K. (1980). Neocognitron: a self organizing neural network<\/span><\/span><\/span><\/span>model for a mechanism of pattern recognition unaffected by shift in po-<\/span><\/span><\/span><\/span>sition. Biol. Cybern. 36, 193\u2013202<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              74. \n

                                                LeCun, Y., and Yoshua, B. (1995). Convolutional networks for images,<\/span><\/span><\/span><\/span>speech, and time series. The Handbook of Brain Theory and Neural Net-<\/span><\/span><\/span><\/span>works 3361, 255\u2013258<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              75. \n

                                                Bao, Y., Sivanandan, S., and Karaletsos, T. (2023). Channel Vision Trans-<\/span><\/span><\/span><\/span>formers: An Image Is Worth c x 16 x 16 WordsThe Twelfth. International<\/span><\/span><\/span><\/span>Conference on Learning Representations 4<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              76. \n

                                                Kraus, O., Kenyon-Dean, K., Saberian, S., Fallah, M., McLean, P., Leung,<\/span><\/span><\/span><\/span>J., Sharma, V., Khan, A., Balakrishnan, J., Celik, S., et al. (2024). Masked<\/span><\/span><\/span><\/span>autoencoders for microscopy are scalable learners of cellular biology. In<\/span><\/span><\/span><\/span>IEEE Conference on Computer Vision and Pattern Recognition (CVPR),pp. 11757\u201311768<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              77. \n

                                                Bao, Y., and Karaletsos, T. (2023). Contextual vision transformers for<\/span><\/span><\/span><\/span>robust representation learning. Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              78. \n

                                                Lopez, R., Regier, J., Cole, M.B., Jordan, M.I., and Yosef, N. (2018). Deep<\/span><\/span><\/span><\/span>generative modeling for single-cell transcriptomics. Nature Methods 15,<\/span><\/span><\/span><\/span>1053\u20131058<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              79. \n

                                                Theodoris, C.V., Xiao, L., Chopra, A., Chaffin, M.D., Al Sayed, Z.R., Hill,<\/span><\/span><\/span><\/span>M.C., Mantineo, H., Brydon, E.M., Zeng, Z., Liu, X.S., et al. (2023). Trans-<\/span><\/span><\/span><\/span>fer learning enables predictions in network biology. Nature 618, 616\u2013624<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              80. \n

                                                Kobayashi-Kirschvink, K.J., Comiter, C.S., Gaddam, S., Joren, T., Grody,<\/span><\/span><\/span><\/span>E.I., Ounadjela, J.R., Zhang, K., Ge, B., Kang, J.W., Xavier, R.J., et al.<\/span><\/span><\/span><\/span>(2024). Prediction of single-cell RNA expression profiles in live cells by<\/span><\/span><\/span><\/span>Raman microscopy with Raman2RNA. Nat. Biotechnol. 42, 1726\u20131734<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              81. \n

                                                Ryu, J., Lopez, R., Bunne, C., and Regev, A. (2024). Cross-modality<\/span><\/span><\/span><\/span>matching and prediction of perturbation responses with labeled<\/span><\/span><\/span><\/span>Gromov-Wasserstein optimal transport. Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              82. \n

                                                Saar, K.L., Scrutton, R.M., Bloznelyte, K., Morgunov, A.S., Good, L.L.,<\/span><\/span><\/span><\/span>Lee, A.A., Teichmann, S.A., and Knowles, T.P.J. (2024). Protein Conden-<\/span><\/span><\/span><\/span>sate Atlas from predic- tive models of heteromolecular condensate<\/span><\/span><\/span><\/span>composition. Nat. Commun. 15, 5418<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              83. \n

                                                Macosko, E.Z., Basu, A., Satija, R., Nemesh, J., Shekhar, K., Goldman,<\/span><\/span><\/span><\/span>M., Tirosh, I., Bialas, A., Kamitaki, N., Martersteck, E., et al. (2015). Highly<\/span><\/span><\/span><\/span>parallel genome-wide expression profiling of individual cells using nano-<\/span><\/span><\/span><\/span>liter droplets. Cell 161, 1202\u20131214<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              84. \n

                                                Sta\u02da hl, P.L., Salme\u00b4 n, F., Vickovic, S., Lundmark, A., Navarro, J.F., Mag-<\/span><\/span><\/span><\/span>nusson, J., Giacomello, S., Asp, M., Westholm, J.O., Huss, M., et al.<\/span><\/span><\/span><\/span>(2016). Visualization and analysis of gene expression in tissue sections<\/span><\/span><\/span><\/span>by spatial transcriptomics. Science 353, 78\u201382<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              85. \n

                                                Lundberg, E., and Borner, G.H.H. (2019). Spatial proteomics: a powerful<\/span><\/span><\/span><\/span>discovery tool for cell biology. Nat. Rev. Mol. Cell Biol. 20, 285\u2013302<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              86. \n

                                                Marconato, L., Palla, G., Yamauchi, K.A., Virshup, I., Heidari, E., Treis, T., Vierdag, W.M., Toth, M., Stockhaus, S., Shrestha, R.B., et al. (2024). Spa- tialData: an open and universal data framework for spatial omics. Nat. Methods. <\/span><\/span>https:\/\/doi.org\/10.1038\/s41592-024-02212-x<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              87. \n

                                                Somnath, V.R., Pariset, M., Hsieh, Y.-P., Martinez, M.R., Krause, A., and<\/span><\/span><\/span><\/span>Bunne, C. (2023). Aligned Diffusion Schro\u00a8 dinger Bridges. In Uncertainty<\/span><\/span><\/span><\/span>in Artificial Intelligence, pp. 1985\u20131995<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              88. \n

                                                Katharopoulos, A., Vyas, A., Pappas, N., and Fleuret, F. (2020). Fast au-<\/span><\/span><\/span><\/span>toregressive transformers with linear attention. In International Confer-<\/span><\/span><\/span><\/span>ence on Machine Learning<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              89. \n

                                                Abramson, J., Adler, J., Dunger, J., Evans, R., Green, T., Pritzel, A., Ron-<\/span><\/span><\/span><\/span>neberger, O., Willmore, L., Ballard, A.J., Bambrick, J., et al. (2024). Accu-<\/span><\/span><\/span><\/span>rate structure prediction of biomolecular interactions with AlphaFold 3.<\/span><\/span><\/span><\/span>Nature 630, 493\u2013500<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              90. \n

                                                Norman, T.M., Horlbeck, M.A., Replogle, J.M., Ge, A.Y., Xu, A., Jost, M.,<\/span><\/span><\/span><\/span>Gilbert, L.A., and Weissman, J.S. (2019). Exploring genetic interaction<\/span><\/span><\/span><\/span>manifolds constructed from rich single-cell phenotypes. Science 365,<\/span><\/span><\/span><\/span>786\u2013793<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              91. \n

                                                Lawson, M.J., Camsund, D., Larsson, J., Baltekin, O<\/span><\/span><\/span>\u00a8 <\/span><\/span><\/span>., Fange, D., and Elf,J. (2017). In situ genotyping of a pooled strain library after characterizing<\/span><\/span><\/span><\/span><\/span>complex phenotypes. Mol. Syst. Biol. 13, 947<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              92. \n

                                                Papamarkou, T., Skoularidou, M., Palla, K., Aitchison, L., Arbel, J., Dun-<\/span><\/span><\/span><\/span>son, D., Filliponne, M., Fortuin, V., Hennig, P., Hernandez-Lobato, J.M.,<\/span><\/span><\/span><\/span>et al. (2024). Position: bayesian deep learning is needed in the age of<\/span><\/span><\/span><\/span>large-scale AI. In Forty-First International Conference on Machine<\/span><\/span><\/span><\/span>Learning<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              93. \n

                                                D\u2019Angelo, F., and Fortuin, V. (2021). Wenzel F.On Stein Variational Neural<\/span><\/span><\/span><\/span>Network. Ensembles Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              94. \n

                                                Ober, S.W., Rasmussen, C.E., and van der Wilk, M. (2021). The promises<\/span><\/span><\/span><\/span>and pitfalls of deep kernel learning. In Conference on Uncertainty in Arti-<\/span><\/span><\/span><\/span>ficial Intelligence, pp. 1206\u20131216<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              95. \n

                                                Karaletsos, T. (2020). Bui T.D.Hierarchical Gaussian Process Priors for<\/span><\/span><\/span><\/span>Bayesian Neural Network Weights. Adv. Neural Inf. Process. Syst. 33,<\/span><\/span><\/span><\/span>17141\u201317152<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              96. \n

                                                Kapoor, S., Maddox, W.J., Izmailov, P., and Wilson, A.G. (2022). On un-<\/span><\/span><\/span><\/span>certainty, tempering, and data augmentation in bayesian classification.<\/span><\/span><\/span><\/span>Adv. Neural Inf. Process. Syst. 35, 18211\u201318225<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              97. \n

                                                Lakshminarayanan, B., Pritzel, A., and Blundell, C. (2017). Simple and<\/span><\/span><\/span><\/span>Scalable Predictive Uncertainty Estimation using Deep Ensembles.<\/span><\/span><\/span><\/span>Adv. Neural Inf. Process. Syst. 30<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              98. \n

                                                Angelopoulos, A.N., and Bates, S. (2021). A gentle introduction to<\/span><\/span><\/span><\/span>conformal prediction and distribution-free uncertainty quantification.<\/span><\/span><\/span><\/span>Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              99. \n

                                                Cherian, J.J., Gibbs, I., and Cande` s, E.J. (2024). Large language model<\/span><\/span><\/span><\/span>validity via enhanced conformal prediction methods. Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              100. \n

                                                Cho, N.H., Cheveralls, K.C., Brunner, A.D., Kim, K., Michaelis, A.C., Ra-ghavan, P., Kobayashi, H., Savy, L., Li, J.Y., Canaj, H., et al. (2022). Open-<\/span><\/span><\/span><\/span>Cell: endogenous tagging for the cartography of human cellular organiza-<\/span><\/span><\/span><\/span>tion. Science 375, eabi6983<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              101. \n

                                                Uhle\u00b4 n, M., Fagerberg, L., Hallstro\u00a8 m, B.M., Lindskog, C., Oksvold, P.,<\/span><\/span><\/span><\/span>Mardinoglu, A., Sivertsson, A<\/span><\/span><\/span>\u02da <\/span><\/span><\/span>., Kampf, C., Sjo\u00a8 stedt, E., Asplund, A.,<\/span><\/span><\/span><\/span><\/span>et al.. (2015). Proteomics. Tissue-based map of the human proteome.<\/span><\/span><\/span><\/span>Science 347, 1260419<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              102. \n

                                                Berger, C., Premaraj, N., Ravelli, R.B.G., Knoops, K., Lo\u00b4 pez-Iglesias, C.,<\/span><\/span><\/span><\/span>and Peters, P.J. (2023). Cryo-electron tomography on focused ion beam<\/span><\/span><\/span><\/span>lamellae transforms structural cell biology. Nat. Methods 20, 499\u2013511<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              103. \n

                                                Loconte, V., Chen, J.H., Vanslembrouck, B., Ekman, A.A., McDermott,<\/span><\/span><\/span><\/span>G., Le Gros, M.A., and Larabell, C.A. (2023). Soft X-ray tomograms pro-<\/span><\/span><\/span><\/span>vide a structural basis for whole-cell modeling. FASEB J. 37, e22681<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              104. \n

                                                Moffitt, J.R., Lundberg, E., and Heyn, H. (2022). The emerging landscape<\/span><\/span><\/span><\/span>of spatial profiling technologies. Nat. Rev. Genet. 23, 741\u2013759<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              105. \n

                                                Vandereyken, K., Sifrim, A., Thienpont, B., and Voet, T. (2023). Methods<\/span><\/span><\/span><\/span>and applications for single-cell and spatial multi-omics. Nat. Rev. Genet.<\/span><\/span><\/span><\/span>24, 494\u2013515<\/span><\/span><\/em><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              106. \n

                                                Tabula Sapiens Consortium*, <\/span><\/span><\/span>Jones, R.C., Karkanias, J., Krasnow, M.A.,<\/span><\/span><\/span><\/span><\/span>Pisco, A.O., Quake, S.R., Salzman, J., Yosef, N., Bulthaup, B., Brown, P.,<\/span><\/span><\/span><\/span>et al. (2022). The tabula sapiens: A multiple-organ, single-cell transcrip-<\/span><\/span><\/span><\/span>tomic atlas of humans. Science 376, eabl4896<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              107. \n

                                                He, B., Bergenstra\u02da hle, L., Stenbeck, L., Abid, A., Andersson, A., Borg, A<\/span><\/span><\/span>\u02da <\/span><\/span><\/span>.,<\/span><\/span><\/span><\/span><\/span>Maaskola, J., Lundeberg, J., and Zou, J. (2020). Integrating spatial gene<\/span><\/span><\/span><\/span>expression and breast tumour morphology via deep learning. Nat. Bio-<\/span><\/span><\/span><\/span>med. Eng. 4, 827\u2013834<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              108. \n

                                                Bock, C., Boutros, M., Camp, J.G., Clarke, L., Clevers, H., Knoblich, J.A.,<\/span><\/span><\/span><\/span>Liberali, P., Regev, A., Rios, A.C., Stegle, O., et al. (2021). The organoid<\/span><\/span><\/span><\/span>cell atlas. Nat. Biotechnol. 39, 13\u201317<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              109. \n

                                                Tabula; Muris Consortium; Overall coordination; Logistical coordination;<\/span><\/span><\/span><\/span>Organ collection and processing; Library preparation and sequencing;<\/span><\/span><\/span><\/span>Computational data analysis; Cell type annotation; Writing group; Sup-<\/span><\/span><\/span><\/span>plemental text writing group; Principal investigators (2018). Single-cell<\/span><\/span><\/span><\/span>transcriptomics of 20 mouse organs creates a tabula muris. Nature<\/span><\/span><\/span><\/span>562, 367\u2013372<\/span><\/span><\/em><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              110. \n

                                                Li, H., Janssens, J., De Waegeneer, M., Kolluru, S.S., Davie, K., Gardeux,<\/span><\/span><\/span><\/span>V., Saelens, W., David, F.P.A., Brbi<\/span><\/span><\/span>c\u00b4<\/span><\/span><\/span>, M., Spanier, K., et al. (2022). Fly Cell<\/span><\/span><\/span><\/span><\/span>Atlas: A single-nucleus transcriptomic atlas of the adult fruit fly. Science<\/span><\/span><\/span><\/span>375, eabk2432<\/span><\/span><\/em><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              111. \n

                                                Lange, M., Granados, A., Vijaykumar, S., Bragantini, J., Ancheta, S., San-<\/span><\/span><\/span><\/span>thosh, S., Borja, M., Kobayashi, H., McGeever, E., Solak, A.C., et al.<\/span><\/span><\/span><\/span>(2023). Zebrahub \u2013 Multimodal zebrafish Developmental Atlas Reveals<\/span><\/span><\/span><\/span>the State Transition Dynamics of Late Vertebrate Pluripotent Axial Pro-<\/span><\/span><\/span><\/span>genitors. Preprint at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              112. \n

                                                Katz, K., Shutov, O., Lapoint, R., Kimelman, M., Brister, J.R., and O\u2019Sul-<\/span><\/span><\/span><\/span>livan, C. (2022). The Sequence Read Archive: a decade more of explosive<\/span><\/span><\/span><\/span>growth. Nucleic Acids Res. 50, D387\u2013D390<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              113. \n

                                                Achiam, J., et al. (2023). GPT-4 technical report. Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              114. \n

                                                Ding, F., and Steinhardt, J.N. (2024). Protein language models are biased<\/span><\/span><\/span><\/span>by unequal sequence sampling across the tree of life. Preprint at bioRxiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              115. \n

                                                Liao, W.-W., Asri, M., Ebler, J., Doerr, D., Haukness, M., Hickey, G., Lu,<\/span><\/span><\/span><\/span>S., Lucas, J.K., Monlong, J., Abel, H.J., et al. (2023). A draft human pan-<\/span><\/span><\/span><\/span>genome reference. Nature 617, 312\u2013324<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              116. \n

                                                Liu, J., Shen, Z., He, Y., Zhang, X., Xu, R., Yu, H., and Cui, P. (2021). To-wards out-of-distribution generalization: A survey. Preprint at arXiv<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              117. \n

                                                Nisonoff, H., Wang, Y., and Listgarten, J. (2023). Coherent blend- ing of biophysics-based knowledge with bayesian neural networks for robust protein property prediction. ACS Synth. Biol. <\/span><\/span>12<\/span><\/span><\/em>, 3242\u20133251. <\/span><\/span>https:\/\/<\/span><\/span><\/span><\/span><\/span>doi.org\/10.1021\/acssynbio.3c00217<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              118. \n

                                                Zheng, F., Kelly, M.R., Ramms, D.J., Heintschel, M.L., Tao, K., Tutuncuo-<\/span><\/span><\/span><\/span>glu, B., Lee, J.J., Ono, K., Foussard, H., Chen, M., et al. (2021). Interpre-<\/span><\/span><\/span><\/span>tation of cancer mutations using a multiscale map of protein systems.<\/span><\/span><\/span><\/span>Science 374, eabf3067<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              119. \n

                                                Ma, J., Yu, M.K., Fong, S., Ono, K., Sage, E., Demchak, B., Sharan, R.,<\/span><\/span><\/span><\/span>and Ideker, T. (2018). Using deep learning to model the hierarchical struc-<\/span><\/span><\/span><\/span>ture and function of a cell. Nat. Methods 15, 290\u2013298<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              120. \n

                                                Gao, S., Fang, A., Huang, Y., Giunchiglia, V., Noori, A., Schwarz, J.R., Ek-<\/span><\/span><\/span><\/span>tefaie, Y., Kondic, J., and Zitnik, M. (2024). Empowering biomedical dis-<\/span><\/span><\/span><\/span>covery with AI agents. Cell 187, 6125\u20136151<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              121. \n

                                                Hurrell, T., Naidoo, J., Ntlhafu, T., and Scholefield, J. (2024). An African<\/span><\/span><\/span><\/span>perspective on genetically diverse human induced pluripotent stem cell<\/span><\/span><\/span><\/span>lines. Nat. Commun. 15, 8581<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              122. \n

                                                Gibbs, R.A., Belmont, J.W., Hardenbol, P., Willis, T.D., Yu, F.L., Yang,H.M., Ch\u2019ang, L.Y., Huang, W., Shen, B., Tam, Y., et al. (2003). The inter-<\/span><\/span><\/span><\/span>national HapMap project. Nature 5, 467\u2013475<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              123. \n

                                                Cancer; Genome; Atlas; Research Network, Weinstein, J.N., Collisson,<\/span><\/span><\/span><\/span>E.A., Mills, G.B., Shaw, K.R.M., Ozenberger, B.A., Ellrott, K., Shmulevich,I., Sander, C., and Stuart, J.M. (2013). The cancer genome atlas pan-can-<\/span><\/span><\/span><\/span>cer analysis project. Nat. Genet. 45, 1113\u20131120<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              124. \n

                                                ENCODE Project Consortium (2012). An integrated encyclopedia of DNA<\/span><\/span><\/span><\/span>elements in the human genome. Nature 489, 57\u201374<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              125. \n

                                                Lonsdale, J., Thomas, J., Salvatore, M., Phillips, R., Lo, E., Shad, S.,<\/span><\/span><\/span><\/span>Hasz, R., Walters, G., Garcia, F., Young, N., and Foster, B. (2013). The<\/span><\/span><\/span><\/span>Genotype-Tissue Expression (GTEx) project. Nat. Genet. 45, 580\u2013585<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              126. \n

                                                Ponte\u00b4 n, F., Jirstro\u00a8 m, K., and Uhlen, M. (2008). The Human Protein Atlas\u2013a<\/span><\/span><\/span><\/span>tool for pathology. J. Pathol. 216, 387\u2013393<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              127. \n

                                                Downey, P., and Peakman, T.C. (2008). Design and implementation of a<\/span><\/span><\/span><\/span>high-throughput biological sample processing facil- ity using modern<\/span><\/span><\/span><\/span>manufacturing principles. Int. J. Epidemiol. 37 (Suppl 1 ), i46\u2013i50<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              128. \n

                                                Madani, A., Krause, B., Greene, E.R., Subramanian, S., Mohr, B.P., Hol-<\/span><\/span><\/span><\/span>ton, J.M., Olmos, J.L., Xiong, C., Sun, Z.Z., Socher, R., et al. (2023). Large<\/span><\/span><\/span><\/span>language models generate func- tional protein sequences across diverse<\/span><\/span><\/span><\/span>families. Nat. Biotechnol. 41, 1099\u20131106<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              129. \n

                                                Nelson, M.R., Tipney, H., Painter, J.L., Shen, J., Nicoletti, P., Shen, Y.,<\/span><\/span><\/span><\/span>Floratos, A., Sham, P.C., Li, M.J., Wang, J., et al. (2015). The support<\/span><\/span><\/span><\/span>of human genetic evidence for approved drug indications. Nat. Genet.<\/span><\/span><\/span><\/span>47, 856\u2013860<\/span><\/span><\/em><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              130. \n

                                                Mason, C., Brindley, D.A., Culme-Seymour, E.J., and Davie, N.L. (2011).<\/span><\/span><\/span><\/span>Cell therapy industry: billion dollar global business with unlimited poten-<\/span><\/span><\/span><\/span>tial. Regen. Med. 6, 265\u2013272<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              131. \n

                                                Bashor, C.J., Hilton, I.B., Bandukwala, H., Smith, D.M., and Veiseh, O.<\/span><\/span><\/span><\/span>(2022). Engineering the next generation of cell-based therapeutics. Nat.<\/span><\/span><\/span><\/span>Rev. Drug Discov. 21, 655\u2013675<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              132. \n

                                                Jia, Q., Wang, A., Yuan, Y., Zhu, B., and Long, H. (2022). Heterogeneity of<\/span><\/span><\/span><\/span>the tumor immune microenvironment and its clinical relevance. Exp.<\/span><\/span><\/span><\/span>Hematol. Oncol. 11, 24<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              133. \n

                                                Melssen, M.M., Sheybani, N.D., Leick, K.M., and Slingluff, C.L. (2023).<\/span><\/span><\/span><\/span>Barriers to immune cell infiltration in tumors. J. Immunother. Cancer 11<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              134. \n

                                                Chow, A., Perica, K., Klebanoff, C.A., and Wolchok, J.D. (2022). Clinical<\/span><\/span><\/span><\/span>implications of T cell exhaustion for cancer immunotherapy. Nature Re-<\/span><\/span><\/span><\/span>views Clinical Oncology 19, 775\u2013790<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              135. \n

                                                de Visser, K.E., and Joyce, J.A. (2023). The evolving tumor microenviron-<\/span><\/span><\/span><\/span>ment: from cancer initiation to metastatic outgrowth. Cancer Cell 41,<\/span><\/span><\/span><\/span>374\u2013403<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              136. \n

                                                Barkley, D., Moncada, R., Pour, M., Liberman, D.A., Dryg, I., Werba, G.,<\/span><\/span><\/span><\/span>Wang, W., Baron, M., Rao, A., Xia, B., et al. (2022). Cancer cell states<\/span><\/span><\/span><\/span>recur across tumor types and form specific interactions with the tumor<\/span><\/span><\/span><\/span>microenvironment. Nat. Genet. 54, 1192\u20131201<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              137. \n

                                                Schwartzberg, L., Kim, E.S., Liu, D., and Schrag, D. (2017). Precision<\/span><\/span><\/span><\/span>oncology: who, how, what, when, and when not? American Society of<\/span><\/span><\/span><\/span>Clinical Oncology Educational Book 37, 160\u2013169<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                              138. \n

                                                Aebersold, R., Agar, J.N., Amster, I.J., Baker, M.S., Bertozzi, C.R., Boja,<\/span><\/span><\/span><\/p>\n<\/li>\n<\/ol>\n

                                                E.S., Costello, C.E., Cravatt, B.F., Fenselau, C., Garcia, B.A., et al. (2018).<\/span><\/span><\/span><\/span>How many human proteoforms are there? Nat. Chem. Biol. 14, 206\u2013214<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n

                                                  \n
                                                1. \n

                                                  Katsoulakis, E., Wang, Q., Wu, H., Shahriyari, L., Fletcher, R., Liu, J.,<\/span><\/span><\/span><\/span>Achenie, L., Liu, H., Jackson, P., Xiao, Y., et al. (2024). Digital twins for<\/span><\/span><\/span><\/span>health: a scoping review. npj Digit. Med. 7, 77<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                2. \n

                                                  Rajewsky, N., Almouzni, G., Gorski, S.A., Aerts, S., Amit, I., Bertero,<\/span><\/span><\/span><\/span>M.G., Bock, C., Bredenoord, A.L., Cavalli, G., Chiocca, S., et al. (2020).<\/span><\/span><\/span><\/span>Lifetime and improving European healthcare through cell-based inter-<\/span><\/span><\/span><\/span>ceptive medicine. Nature 587, 377\u2013386<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                3. \n

                                                  Alix-Panabie` res, C., and Pantel, K. (2021). Liquid biopsy: from discovery<\/span><\/span><\/span><\/span>to clinical application. Cancer Discov. 11, 858\u2013873<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                4. \n

                                                  Vaishnav, E.D., de Boer, C.G., Molinet, J., Yassour, M., Fan, L., Adiconis,<\/span><\/span><\/span><\/span>X., Thompson, D.A., Levin, J.Z., Cubillos, F.A., and Regev, A. (2022). The<\/span><\/span><\/span><\/span>evolution, evolvability and engineering of gene regulatory DNA. Nature<\/span><\/span><\/span><\/span>603, 455\u2013463<\/span><\/span><\/em><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                5. \n

                                                  Go\u00b4 mez-de-Mariscal, E., Garc\u0131\u00b4a-Lo\u00b4 pez-de-Haro, C., Ouyang, W., Donati,<\/span><\/span><\/span><\/span>L., Lundberg, E., Unser, M., Mun\u02dc oz-Barrutia, A., and Sage, D. (2021).<\/span><\/span><\/span><\/span>DeepImageJ: A user-friendly environment to run deep learning models<\/span><\/span><\/span><\/span>in ImageJ. Nat. Methods 18, 1192\u20131195<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                6. \n

                                                  Le, T., Winsnes, C.F., Axelsson, U., Xu, H., Mohanakrishnan Kaimal, J.,<\/span><\/span><\/span><\/span>Mahdessian, D., Dai, S., Makarov, I.S., Ostankovich, V., Xu, Y., et al.<\/span><\/span><\/span><\/span>(2022). Analysis of the human protein atlas weakly supervised single-<\/span><\/span><\/span><\/span>cell classification competition. Nat. Methods 19, 1221\u20131229<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                7. \n

                                                  Chen, R.J., Ding, T., Lu, M.Y., Williamson, D.F.K., Jaume, G., Song, A.H.,<\/span><\/span><\/span><\/span>Chen, B., Zhang, A., Shao, D., Shaban, M., et al. (2024). Towards a gen-<\/span><\/span><\/span><\/span>eral-purpose foundation model for computational pathology. Nat. Med.<\/span><\/span><\/span><\/span>30, 850\u2013862<\/span><\/span><\/em><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                8. \n

                                                  Moen, E., Bannon, D., Kudo, T., Graf, W., Covert, M., and Van Valen, D.<\/span><\/span><\/span><\/span>(2019). Deep learning for cellular image analysis. Nat. Methods 16,<\/span><\/span><\/span><\/span>1233\u20131246<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                9. \n

                                                  Avsec,<\/span><\/span><\/span> Z\u02c7<\/span><\/span><\/span>., Weilert, M., Shrikumar, A., Krueger, S., Alexandari, A., Dalal,<\/span><\/span><\/span><\/span><\/span>K., Fropf, R., McAnany, C., Gagneur, J., Kundaje, A., et al. (2021).<\/span><\/span><\/span><\/span>Base-resolution models of transcription-factor binding reveal soft motif<\/span><\/span><\/span><\/span>syntax. Nat. Genet. 53, 354\u2013366<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                10. \n

                                                  Ho, J., Jain, A., and Abbeel, P. (2020). Denoising diffusion proba- bilistic<\/span><\/span><\/span><\/span>models. Adv. Neural Inf. Process. Syst. 33, 6840\u20136851<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                11. \n

                                                  Lipman, Y., Chen, R.T., Ben-Hamu, H., Nickel, M., and Le, M. (2023).<\/span><\/span><\/span><\/span>Flow Matching for Generative Modeling. International Conference on<\/span><\/span><\/span><\/span>Learning Representations<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                12. \n

                                                  Scarselli, F., Gori, M., Tsoi, A.C., Hagenbuchner, M., and Monfardini, G.<\/span><\/span><\/span><\/span>(2009). The graph neural network model. IEEE Trans. Neural Netw. 20,<\/span><\/span><\/span><\/span>61\u201380<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                13. \n

                                                  Cao, Y., and Shen, Y. (2020). Energy-based graph convolutional net-<\/span><\/span><\/span><\/span>works for scoring protein docking models. Proteins 88, 1091\u20131099<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                14. \n

                                                  Brbi<\/span><\/span><\/span>c\u00b4<\/span><\/span><\/span>, M., Cao, K., Hickey, J.W., Tan, Y., Snyder, M.P., Nolan, G.P., and<\/span><\/span><\/span><\/span><\/span>Leskovec, J. (2022). Annotation of spatially resolved single-cell data with<\/span><\/span><\/span><\/span>STELLAR. Nat. Methods 19, 1411\u20131418<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                15. \n

                                                  Wu, Z., Trevino, A.E., Wu, E., Swanson, K., Kim, H.J., D\u2019Angio, H.B., Pre-<\/span><\/span><\/span><\/span>ska, R., Charville, G.W., Dalerba, P.D., Egloff, A.M., et al. (2022). Graph<\/span><\/span><\/span><\/span>deep learning for the characterization of tumour microenvironments<\/span><\/span><\/span><\/span>from spatial protein profiles in tissue specimens. Nat. Biomed. Eng. 6,<\/span><\/span><\/span><\/span>1435\u20131448<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n

                                                16. \n

                                                  Hamilton, W., Ying, Z., and Leskovec, J. (2017). Inductive Representation<\/span><\/span><\/span><\/span>Learning on Large Graphs. Adv. Neural Inf. Process. Syst. 30<\/span><\/span><\/span>.<\/span><\/span><\/span><\/p>\n<\/li>\n<\/ol>\n<\/section>\n<\/section>\n


                                                  <\/span><\/span><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

                                                  \n
                                                  \n

                                                  \u53c2\u8003\u6587\u732e\u53ef<\/span>\u4e0a\u4e0b\u6ed1\u52a8<\/span><\/strong>\u67e5\u770b<\/span><\/p>\n<\/section>\n<\/section>\n


                                                  <\/span><\/span><\/strong><\/span><\/section>\n

                                                  <\/span><\/span><\/strong><\/span><\/section>\n
                                                  \n
                                                  \n
                                                  \n
                                                  \n

                                                  \u751f\u547d\u590d\u6742\u6027\u8bfb\u4e66\u4f1a\uff1a<\/span><\/strong><\/p>\n

                                                  \u751f\u547d\u590d\u6742\u7cfb\u7edf\u7684\u6784\u6210\u539f\u7406<\/span><\/strong>
                                                  <\/span><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n


                                                  <\/span><\/section>\n

                                                  \u5728\u751f\u7269\u5b66\u4e2d\u5fc3\u6cd5\u5219\u7684\u8d77\u70b9\uff0c\u57fa\u56e0\u4f5c\u4e3a\u751f\u547d\u590d\u6742\u7cfb\u7edf\u7684\u9057\u4f20\u4fe1\u606f\u8f7d\u4f53\uff0c\u5728\u751f\u547d\u5468\u671f\u5185\u7a33\u5b9a\u5b58\u5728\uff1b\u800c\u4f4d\u4e8e\u4e2d\u5fc3\u6cd5\u5219\u672b\u7aef\u7684\u86cb\u767d\u8d28\uff0c\u5176\u7ec4\u7ec7\u6784\u6210\u548c\u65f6\u7a7a\u53d8\u5316\u7684\u590d\u6742\u6027\u5448\u6307\u6570\u5f0f\u589e\u957f\u3002\u968f\u7740\u5206\u5b50\u751f\u7269\u5b66\u6570\u5341\u5e74\u6765\u7684\u7a81\u98de\u731b\u8fdb\uff0c\u5c24\u5176\u662f\u751f\u547d\u7ec4\u5b66\uff08\u57fa\u56e0\u7ec4\u5b66\u3001\u8f6c\u5f55\u7ec4\u5b66\u3001\u86cb\u767d\u8d28\u7ec4\u5b66\u548c\u4ee3\u8c22\u7ec4\u5b66\u7b49\u7684\u96c6\u5408\uff09\u7b49\u9886\u57df\u7684\u65e5\u65b0\u6708\u5f02\uff0c\u5f53\u4ee3\u751f\u547d\u79d1\u5b66\u4e34\u8fd1\u7206\u53d1\u7684\u8fb9\u7f18\u3002\u5982\u6b64\u6d77\u91cf\u7684\u6570\u636e\u5982\u4f55\u5e2e\u52a9\u6211\u4eec\u63ed\u793a\u5b87\u5b99\u4e2d\u6700\u590d\u6742\u7684\u7269\u8d28\u7cfb\u7edf\u2014\u2014\u201c\u4eba\u4f53\u201d\u7684\u6784\u6210\u539f\u7406\u548c\u8bbe\u8ba1\u539f\u7406\uff1f\u9610\u91ca\u4eba\u7c7b\u53d1\u80b2\u3001\u8870\u8001\u548c\u91cd\u5927\u75be\u75c5\u7684\u53d1\u751f\u673a\u5236\uff1f<\/span><\/span><\/p>\n


                                                  <\/span><\/span><\/p>\n

                                                  \u96c6\u667a\u4ff1\u4e50\u90e8\u8054\u5408\u897f\u6e56\u5927\u5b66\u7406\u5b66\u9662\u53ca\u4ea4\u53c9\u79d1\u5b66\u4e2d\u5fc3\u8bb2\u5e2d\u6559\u6388\u6c64\u96f7\u7ff0\uff0c\u56fd\u5bb6\u86cb\u767d\u8d28\u79d1\u5b66<\/span><\/span>\u4e2d\u5fc3\uff08\u5317\u4eac\uff09\u526f\u7814\u7a76\u5458\u5e38\u4e58\u3001\u674e\u6768\uff0c\u9999\u6e2f\u6d78\u4f1a\u5927\u5b66\u52a9\u7406\u6559\u6388\u5510\u4e7e\u5143\uff0c\u5317\u4eac\u5927\u5b66\u524d\u6cbf\u4ea4\u53c9\u5b66\u79d1\u7814\u7a76\u9662\u7814\u7a76\u5458\u6797\u4e00\u701a\uff0c\u4e2d\u56fd\u79d1\u5b66\u9662\u5206\u5b50\u7ec6\u80de\u79d1\u5b66\u5353\u8d8a\u521b\u65b0\u4e2d\u5fc3\u535a\u58eb\u540e\u5510\u8bd7\u5a55\uff0c\u5171\u540c\u53d1\u8d77<\/span><\/span>\u300c\u751f\u547d\u590d\u6742\u6027\uff1a\u751f\u547d\u590d\u6742\u7cfb\u7edf\u7684\u6784\u6210\u539f\u7406\u300d\u8bfb\u4e66\u4f1a<\/a><\/span>\uff0c\u4ece\u5fae\u89c2\u7ec6\u80de\u5c3a\u5ea6\u3001\u4ecb\u89c2\u7ec4\u7ec7\u5668\u5b98\u5c3a\u5ea6\u5230\u5b8f\u89c2\u4eba\u4f53\u5c3a\u5ea6\uff0c\u68b3\u7406\u751f\u547d\u79d1\u5b66\u9886\u57df\u4e2d\u7684\u91cd\u8981\u95ee\u9898\u53ca\u91cd\u8981\u6570\u636e\uff0c\u7531\u751f\u7269\u5b66\u5bb6\u63d0\u95ee\uff0c\u5e0c\u671b\u4fc3\u8fdb\u7edf\u8ba1\u7269\u7406\u3001\u673a\u5668\u5b66\u4e60\u65b9\u6cd5\u7814\u7a76\u8005\u548c\u751f\u547d\u79d1\u5b66\u7814\u7a76\u8005\u4e4b\u95f4\u7684\u6df1\u5ea6\u4ea4\u6d41\uff0c\u5efa\u7acb\u8de8\u5b66\u79d1\u5408\u4f5c\u5173\u7cfb\uff0c\u6fc0\u53d1\u65b0\u7684\u7814\u7a76\u601d\u8def\u548c\u5408\u4f5c\u9879\u76ee\u3002\u8bfb\u4e66\u4f1a\u76ee\u524d\u5171\u8fdb\u884c10\u671f\uff0c\u73b0\u5728\u62a5\u540d\u53c2\u4e0e\u8bfb\u4e66\u4f1a\u53ef\u4ee5\u52a0\u5165\u8bfb\u4e66\u4f1a\u793e\u7fa4\uff0c\u89c2\u770b\u89c6\u9891\u56de\u653e\uff0c\u89e3\u9501\u5b8c\u6574\u8bfb\u4e66\u4f1a\u6743\u9650\u3002<\/span><\/span><\/p>\n


                                                  <\/span><\/span><\/section>\n
                                                  \"\u56fe\u7247\"<\/span><\/a><\/span><\/strong><\/span><\/section>\n

                                                  <\/span><\/span><\/section>\n
                                                  \u8be6\u60c5\u8bf7\u89c1\uff1a<\/span><\/span><\/section>\n
                                                  \u751f\u547d\u590d\u6742\u6027\u8bfb\u4e66\u4f1a\uff1a\u4ece\u5fae\u89c2\u5230\u5b8f\u89c2\uff0c\u591a\u5c3a\u5ea6\u89c6\u89d2\u63a2\u7d22\u751f\u547d\u590d\u6742\u7cfb\u7edf\u7684\u6784\u6210\u539f\u7406<\/a><\/span><\/span>
                                                  <\/span><\/section>\n

                                                  <\/span><\/section>\n
                                                  \u63a8\u8350\u9605\u8bfb<\/span><\/span><\/strong><\/strong><\/strong><\/section>\n
                                                  <\/strong><\/section>\n
                                                  1. \u6570\u636e\u9a71\u52a8AI\u865a\u62df\u7ec6\u80de\u63a2\u7d22\uff1a\u4eceArc\u6311\u6218\u8d5b\u770b\u201c\u6570\u636e\u57fa\u5ea7\u201d\u7684\u529b\u91cf<\/span><\/a><\/span><\/strong>
                                                  <\/span><\/strong><\/span><\/section>\n
                                                  2. \u524d\u6cbf\u7efc\u8ff0\uff1a\u4eba\u5de5\u7ec6\u80de\u7684\u8fc7\u53bb\u3001\u73b0\u5728\u3001\u672a\u6765<\/span><\/a><\/span><\/strong>
                                                  <\/span><\/strong><\/span><\/section>\n
                                                  3. \u4ece\u751f\u547d\u5230\u4eba\u5de5\u751f\u547d\uff1a\u672a\u6765\u751f\u547d\u5f62\u6001\u4f1a\u662f\u600e\u6837\uff1f<\/span><\/a><\/span><\/strong><\/section>\n
                                                  <\/strong><\/section>\n
                                                  4. <\/span>\u6d8c\u73b0\u52a8\u529b\u5b66\u5982\u4f55\u7528\u6765\u5206\u6790\u590d\u6742\u7cfb\u7edf\uff1f | \u65b0\u8bfe\u4e0a\u7ebf<\/span><\/a><\/span><\/strong><\/span>
                                                  <\/span><\/strong><\/strong><\/span><\/section>\n
                                                  <\/strong>5. <\/span><\/strong>\u96c6\u667a\u5b66\u56ed\u7cbe\u54c1\u8bfe\u7a0b\u514d\u8d39\u5f00\u653e\uff0c\u89e3\u9501\u7cfb\u7edf\u79d1\u5b66\u4e0e AI \u65b0\u4e16\u754c<\/span><\/a><\/span><\/strong><\/section>\n

                                                  6. \u9ad8\u8003\u5206\u6570\u53ea\u662f\u5f20\u5165\u573a\u5238\uff0c\u4f60\u7684\u79d1\u7814\u5192\u9669\u5728\u8fd9\u91cc\u542f\u822a\uff01<\/span><\/a><\/span><\/strong><\/p>\n

                                                  7. <\/span><\/span><\/strong><\/strong><\/span>\u91cd\u6574\u5316\u7fa4\u4e0e\u975e\u7ebf\u6027\u7269\u7406\uff0c\u5bfb\u627e\u590d\u6742\u7cfb\u7edf\u8de8\u5c3a\u5ea6\u7684\u5206\u6790\u65b9\u6cd5\u4e28\u65b0\u8bfe\u53d1\u5e03<\/a><\/span><\/strong><\/strong><\/span><\/p>\n


                                                  <\/span><\/p>\n


                                                  <\/span><\/span><\/p>\n

                                                  \u70b9\u51fb\u201c\u9605\u8bfb\u539f\u6587\u201d\uff0c\u62a5\u540d\u8bfb\u4e66\u4f1a<\/span><\/span><\/section>\n<\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

                                                  \u5bfc\u8bed \u201c\u6211\u4eec\u80fd\u5728\u8ba1\u7b97\u673a\u91cc\u6784\u5efa\u4e00\u4e2a\u7ec6\u80de\u5417\uff1f\u201d \u8fd9\u4e00\u6839\u672c\u6027\u95ee\u9898\uff0c\u6b63\u968f\u7740\u4eba\u5de5\u667a\u80fd\u548c\u7ec4\u5b66\u6280\u672f\u7684\u8fdb\u6b65\u800c\u8fce\u6765\u7a81\u7834\u6027\u673a\u9047\u3002\u6700\u8fd1\uff0c\u6743\u5a01\u671f\u520aCell\u53d1\u8868\u4e86\u4e00\u7bc7\u91cd\u78c5\u7efc\u8ff0\uff0c\u6df1\u5165\u63a2\u8ba8\u4e86 \u201c\u4eba\u5de5\u667a\u80fd\u865a\u62df\u7ec6\u80de\u201d\uff08AI Virtual Cell, AIVC\uff09 \u8fd9\u4e00\u524d\u6cbf\u6982\u5ff5\u3002\u8fd9\u7bc7\u6587\u7ae0\u6c47\u96c6\u4e86AI\u4e0e\u751f\u547d\u79d1\u5b66\u9886\u57df\u7684\u9876\u5c16\u667a\u6167\u3002\u4f5c\u8005\u56e2\u961f\u9635\u5bb9…<\/p>\n","protected":false},"author":0,"featured_media":64825,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[],"special":[],"_links":{"self":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/64837"}],"collection":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=64837"}],"version-history":[{"count":0,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/64837\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/media\/64825"}],"wp:attachment":[{"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=64837"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=64837"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=64837"},{"taxonomy":"special","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fspecial&post=64837"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}