{"id":66336,"date":"2026-11-03T16:56:20","date_gmt":"2026-11-03T08:56:20","guid":{"rendered":"https:\/\/www.biocloudservice.com\/wordpress\/?p=66336"},"modified":"2026-11-03T16:56:20","modified_gmt":"2026-11-03T08:56:20","slug":"massnet-xuanjinovo-ai-proteomics-gpu-storage","status":"publish","type":"post","link":"https:\/\/www.biocloudservice.com\/wordpress\/massnet-xuanjinovo-ai-proteomics-gpu-storage\/","title":{"rendered":"30 TB \u539f\u59cb\u8d28\u8c31\u300115.4 \u4ebf\u5f20\u8c31\u56fe\uff1aMassNet \u628a AI \u86cb\u767d\u7ec4\u5b66\u63a8\u5230\u591a GPU \u4e0e\u9ad8\u901f\u5b58\u50a8"},"content":{"rendered":"
\n

\u505a AI \u86cb\u767d\u7ec4\u5b66\u65f6\uff0c\u5f88\u591a\u4eba\u5148\u95ee\u201c\u6a21\u578b\u80fd\u4e0d\u80fd\u88c5\u4e0a\u201d\uff0c\u771f\u6b63\u5f00\u8dd1\u540e\u624d\u53d1\u73b0\uff1aGPU \u53ea\u662f\u5176\u4e2d\u4e00\u73af\u3002\u6570\u767e\u4e07\u5f20 MS\/MS \u8c31\u56fe\u8981\u53cd\u590d\u89e3\u6790\u3001\u968f\u673a\u8bfb\u53d6\u3001\u7ec4\u6279\u3001\u63a8\u7406\u548c\u5199\u56de\u7ed3\u679c\uff1b\u5982\u679c\u6570\u636e\u4ecd\u6563\u843d\u5728\u5927\u91cf mzML\u3001MGF \u6216\u4eea\u5668\u539f\u59cb\u6587\u4ef6\u91cc\uff0c\u8bad\u7ec3\u5361\u5e38\u5e38\u5728\u7b49\u78c1\u76d8\uff0c\u663e\u5b58\u518d\u5927\u4e5f\u5403\u4e0d\u6ee1\u3002<\/p>\n

2026 \u5e74 10 \u6708\uff0c\u300aNature Methods\u300b\u53d1\u8868 MassNet\uff1a\u5b83\u6c47\u603b 648 \u4e2a\u9879\u76ee\u300127,643 \u6b21 LC\u2013MS\/MS \u8fd0\u884c\u548c 35 \u4e2a\u7269\u79cd\uff0c\u5171 15.4 \u4ebf\u5f20\u4e8c\u7ea7\u8d28\u8c31\u56fe\u3002\u6570\u636e\u6e90\u7ea6 30 TB\uff0c\u7ecf\u7edf\u4e00\u641c\u7d22\u3001\u8d28\u63a7\u548c\u5f20\u91cf\u5316\u6574\u7406\u540e\u5f62\u6210 4.45 TB \u7684 AI-ready \u8bed\u6599\u3002\u7814\u7a76\u56e2\u961f\u518d\u7528\u5176\u4e2d 1 \u4ebf\u6761 PSM \u8bad\u7ec3 XuanjiNovo\uff0c\u7528\u4e8e\u4ece\u8c31\u56fe\u76f4\u63a5\u9884\u6d4b\u80bd\u6bb5\u5e8f\u5217\u3002<\/p>\n

\"MassNet\u4e0eXuanjiNovo\u7b97\u529b\u89c4\u5212\uff1a30TB\u539f\u59cb\u8d28\u8c31\u30014.45TB\u8bed\u6599\u548c1\u4ebfPSM\u8bad\u7ec3\"
MassNet \u4e0e XuanjiNovo \u8d44\u6e90\u8def\u5f84\uff1a\u5148\u533a\u5206\u9884\u8bad\u7ec3\u63a8\u7406\u3001\u5e73\u53f0\u5fae\u8c03\u548c\u5927\u89c4\u6a21\u8bad\u7ec3\uff0c\u518d\u914d\u7f6e GPU \u4e0e\u9ad8\u901f\u5de5\u4f5c\u76d8\u3002<\/figcaption><\/figure>\n

\u8fd9\u7bc7\u8bba\u6587\u7684\u91cd\u70b9\uff0c\u4e0d\u53ea\u662f\u201c\u6570\u636e\u66f4\u5927\u201d<\/h2>\n

MassNet \u4e0d\u662f\u628a\u516c\u5171\u6570\u636e\u7b80\u5355\u5806\u5728\u4e00\u8d77\u3002\u56e2\u961f\u7528 FragPipe \u548c Sage \u5206\u522b\u641c\u7d22\uff0c\u5728 1% FDR \u4e0b\u83b7\u5f97\u5019\u9009\u7ed3\u679c\uff0c\u518d\u7edf\u4e00\u5143\u6570\u636e\u3001\u8c31\u5cf0\u3001\u80bd\u6bb5\u3001\u7535\u8377\u548c\u4eea\u5668\u4fe1\u606f\u3002\u6309\u8fd0\u884c\u7edf\u8ba1\uff0c\u53bb\u91cd\u540e\u6709 5.48 \u4ebf\u6761\u76ee\u6807 PSM\uff1b\u8fdb\u4e00\u6b65\u505a\u5168\u5c40 FDR \u4e0e\u53bb\u91cd\u540e\uff0c\u4fdd\u7559\u7ea6 5.01 \u4ebf\u6761 PSM\u3001204 \u4e07\u6761\u72ec\u7279\u80bd\u6bb5\u548c 122,056 \u4e2a\u86cb\u767d\u3002<\/p>\n

\u66f4\u503c\u5f97\u6ce8\u610f\u7684\u662f\u6570\u636e\u5f62\u6001\u3002\u4f5c\u8005\u63d0\u51fa MSDT\uff1a\u903b\u8f91\u5c42\u4f7f\u7528 Parquet \u4fdd\u5b58\u7ed3\u6784\u5316\u8c31\u56fe\u4e0e\u9274\u5b9a\u4fe1\u606f\uff0c\u8bad\u7ec3\u524d\u8fd8\u80fd\u8f6c\u6210 LMDB \u5185\u5b58\u6620\u5c04\u6570\u636e\u5e93\u3002\u8fd9\u6837\u6a21\u578b\u53ef\u968f\u673a\u8bfb\u53d6\u9700\u8981\u7684\u6837\u672c\uff0c\u4e0d\u5fc5\u6bcf\u4e2a epoch \u90fd\u91cd\u65b0\u626b\u63cf\u6587\u672c\u6587\u4ef6\u3002\u5bf9\u5927\u89c4\u6a21\u8bad\u7ec3\u800c\u8a00\uff0c\u9ad8\u901f NVMe\u3001\u5185\u5b58\u6620\u5c04\u548c\u6570\u636e\u5206\u7247\u4e0e GPU \u6570\u91cf\u540c\u6837\u91cd\u8981\u3002<\/p>\n

XuanjiNovo \u4f7f\u7528\u975e\u81ea\u56de\u5f52 Transformer\u3001\u8bfe\u7a0b\u5b66\u4e60\u548c\u7cbe\u786e\u8d28\u91cf\u63a7\u5236\u6a21\u5757\u3002\u8bba\u6587\u5206\u522b\u6bd4\u8f83 3000 \u4e07\u30016500 \u4e07\u548c 1 \u4ebf PSM \u8bad\u7ec3\u89c4\u6a21\uff1b1 \u4ebf\u7248\u672c\u7684\u6c28\u57fa\u9178\u7ea7\u53ec\u56de\u7ea6 0.82\uff0c\u80bd\u6bb5\u7ea7\u53ec\u56de\u7ea6 0.68\uff0c\u9ad8\u4e8e 3000 \u4e07\u7248\u672c\u7ea6 0.75 \u548c 0.54\u3002\u4e0d\u8fc7\uff0c\u6570\u636e\u66f4\u5927\u5e76\u4e0d\u81ea\u52a8\u89e3\u51b3\u8de8\u5e73\u53f0\u504f\u79fb\uff1a\u5728\u90e8\u5206\u65e7\u5e73\u53f0\u6216\u5916\u90e8\u6570\u636e\u4e0a\uff0c\u57fa\u7840\u6a21\u578b\u5e76\u975e\u59cb\u7ec8\u6700\u4f18\uff0c\u968f\u540e\u8fd8\u9700\u7528 MassIVE-KB \u6570\u636e\u5fae\u8c03\u3002<\/p>\n

\u5148\u5224\u65ad\u4f60\u9700\u8981\u63a8\u7406\u3001\u5fae\u8c03\uff0c\u8fd8\u662f\u4ece\u5934\u8bad\u7ec3<\/h2>\n
\n\n\n\n\n\n\n\n\n
\u4efb\u52a1\u5c42\u7ea7<\/th>\n\u5178\u578b\u8f93\u5165<\/th>\n\u4e3b\u8981\u74f6\u9888<\/th>\n\u5408\u7406\u8d77\u70b9<\/th>\n<\/tr>\n<\/thead>\n
\u590d\u7528\u9884\u8bad\u7ec3\u6a21\u578b<\/td>\nMGF \u6216 MSDT\uff0c\u5355\u6279\u9879\u76ee<\/td>\nCUDA \u73af\u5883\u3001\u6279\u5927\u5c0f\u3001\u8c31\u56fe\u8bfb\u53d6<\/td>\n\u5355 GPU \u5148\u8dd1 1 \u4e07\u5f20\u8c31\u56fe\uff0c\u8bb0\u5f55\u663e\u5b58\u4e0e\u541e\u5410<\/td>\n<\/tr>\n
\u591a\u6279\u6b21\u6b63\u5f0f\u63a8\u7406<\/td>\n\u591a\u6587\u4ef6\u3001\u6570\u5341\u4e07\u81f3\u5343\u4e07\u8c31\u56fe<\/td>\nNVMe I\/O\u3001\u4efb\u52a1\u5207\u5206\u3001\u7ed3\u679c\u5408\u5e76<\/td>\n\u591a GPU \u961f\u5217\uff1b\u6bcf\u5361\u72ec\u7acb\u5206\u7247\u4e0e\u65e5\u5fd7<\/td>\n<\/tr>\n
\u5e73\u53f0\u6216\u7269\u79cd\u5fae\u8c03<\/td>\n\u5e26\u53ef\u9760\u6807\u7b7e\u7684\u672c\u5730\u6570\u636e<\/td>\n\u8bad\u7ec3\u7a33\u5b9a\u6027\u3001\u9a8c\u8bc1\u96c6\u9694\u79bb\u3001\u68c0\u67e5\u70b9<\/td>\n\u5148\u505a\u5c0f\u89c4\u6a21\u5bf9\u7167\uff0c\u518d\u51b3\u5b9a GPU \u6570\u548c\u8bad\u7ec3\u8f6e\u6b21<\/td>\n<\/tr>\n
\u91cd\u5efa\u5927\u8bed\u6599\u8bad\u7ec3<\/td>\n3000 \u4e07\u20141 \u4ebf PSM<\/td>\n\u591a GPU \u901a\u4fe1\u3001TB \u7ea7\u5b58\u50a8\u3001\u957f\u671f\u4efb\u52a1<\/td>\n\u5206\u5e03\u5f0f\u8bad\u7ec3\u3001\u65ad\u70b9\u7eed\u8dd1\u4e0e\u7248\u672c\u5316\u6570\u636e\u96c6<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

\u8fd9\u5f20\u8868\u6700\u5173\u952e\u7684\u5224\u65ad\u662f\uff1a\u591a\u6570\u8bfe\u9898\u7ec4\u5e76\u4e0d\u9700\u8981\u590d\u523b 1 \u4ebf PSM \u8bad\u7ec3\u3002\u82e5\u76ee\u6807\u53ea\u662f\u7ed9\u81ea\u5df1\u7684 DDA \u6570\u636e\u505a de novo \u6d4b\u5e8f\uff0c\u5148\u7528\u5b98\u65b9\u68c0\u67e5\u70b9\u63a8\u7406\u66f4\u5408\u7406\uff1b\u53ea\u6709\u5f53\u4eea\u5668\u5e73\u53f0\u3001\u788e\u88c2\u65b9\u5f0f\u6216\u7269\u79cd\u5dee\u5f02\u9020\u6210\u7a33\u5b9a\u504f\u79fb\uff0c\u624d\u503c\u5f97\u8bbe\u8ba1\u5fae\u8c03\u3002\u76f4\u63a5\u4ece\u5934\u8bad\u7ec3\u65e2\u8017\u8d44\u6e90\uff0c\u4e5f\u5bb9\u6613\u628a\u6570\u636e\u6cc4\u6f0f\u3001\u9519\u8bef\u6807\u7b7e\u548c\u6279\u6b21\u504f\u5dee\u653e\u5927\u3002<\/p>\n

\u4e3a\u4ec0\u4e48\u670d\u52a1\u5668\u6bd4\u4e34\u65f6\u5de5\u4f5c\u7ad9\u66f4\u9002\u5408<\/h2>\n

\u7b2c\u4e00\uff0c\u4f9d\u8d56\u94fe\u590d\u6742\u3002\u5b98\u65b9\u4ed3\u5e93\u8981\u6c42 Python 3.10\u3001PyTorch 2.1.0\uff0c\u5e76\u6d89\u53ca CUDA\u3001CuPy\u3001CTC \u89e3\u7801\u6269\u5c55\u548c GCC \u517c\u5bb9\uff1bmacOS \u56e0 CUDA \u4f9d\u8d56\u4e0d\u53d7\u652f\u6301\u3002\u4ed3\u5e93\u540c\u65f6\u63d0\u4f9b CUDA 11 \u4e0e CUDA 12 \u7684 Docker \u955c\u50cf\u3002\u628a\u9a71\u52a8\u3001\u5bb9\u5668\u3001\u6a21\u578b\u6743\u91cd\u548c\u6570\u636e\u76ee\u5f55\u56fa\u5b9a\u4e0b\u6765\uff0c\u6bd4\u5728\u4e2a\u4eba\u7535\u8111\u53cd\u590d\u91cd\u88c5\u66f4\u5bb9\u6613\u590d\u73b0\u3002<\/p>\n

\u7b2c\u4e8c\uff0c\u8d44\u6e90\u9700\u8981\u5206\u5c42\u3002\u9884\u5904\u7406\u4e0e MSDT \u8f6c\u6362\u66f4\u504f CPU\u3001\u5185\u5b58\u548c\u78c1\u76d8\uff0c\u8bad\u7ec3\u4e0e\u63a8\u7406\u4f9d\u8d56 GPU\uff0c\u7ed3\u679c\u6c47\u603b\u53c8\u56de\u5230 CPU\/I\/O\u3002\u670d\u52a1\u5668\u53ef\u628a\u8f6c\u6362\u3001\u8bad\u7ec3\u3001\u8bc4\u4f30\u62c6\u6210\u961f\u5217\uff0c\u907f\u514d\u4e00\u4e2a\u8fdb\u7a0b\u5360\u6ee1\u6574\u673a\u3002\u5b98\u65b9\u8bf4\u660e\u663e\u5b58\u9700\u6c42\u53d7 batch_size<\/code> \u548c n_beams<\/code> \u5171\u540c\u5f71\u54cd\uff0c\u9ed8\u8ba4\u5efa\u8bae n_beams=40<\/code>\u3001refine_iters=3<\/code>\u3001\u4fdd\u7559 800 \u4e2a\u5cf0\uff1b\u76f2\u76ee\u589e\u5927 beam \u4f1a\u76f4\u63a5\u589e\u52a0\u8ba1\u7b97\u91cf\u3002<\/p>\n

\u7b2c\u4e09\uff0c\u5927\u4efb\u52a1\u5fc5\u987b\u53ef\u6062\u590d\u3002\u6a21\u578b\u3001\u8f93\u5165\u5206\u7247\u3001\u914d\u7f6e\u3001\u6821\u9a8c\u548c\u3001\u65e5\u5fd7\u548c\u68c0\u67e5\u70b9\u90fd\u5e94\u72ec\u7acb\u4fdd\u5b58\u3002\u8bba\u6587\u7ea7\u7ed3\u679c\u8fd8\u8981\u8bb0\u5f55\u8f6f\u4ef6\u7248\u672c\u3001FDR \u89c4\u5219\u3001\u6570\u636e\u62c6\u5206\u548c\u5916\u90e8\u9a8c\u8bc1\uff0c\u4e0d\u80fd\u53ea\u7559\u4e0b\u4e00\u4e2a\u9884\u6d4b\u8868\u3002<\/p>\n

\u5b58\u50a8\u4e5f\u4e0d\u5e94\u53ea\u6309\u201c\u8f93\u5165\u6587\u4ef6\u6709\u591a\u5927\u201d\u8d2d\u4e70\u3002\u4e00\u4e2a\u7a33\u59a5\u7684\u9879\u76ee\u81f3\u5c11\u5206\u4e09\u5c42\uff1a\u539f\u59cb\u6587\u4ef6\u53ea\u8bfb\u4fdd\u5b58\uff1bMSDT\u3001LMDB \u548c\u8bad\u7ec3\u5206\u7247\u653e\u5728\u9ad8\u901f\u5de5\u4f5c\u76d8\uff1b\u7ed3\u679c\u3001\u65e5\u5fd7\u548c\u68c0\u67e5\u70b9\u8fdb\u5165\u957f\u671f\u76ee\u5f55\u3002\u8f6c\u6362\u671f\u95f4\u539f\u59cb\u6570\u636e\u4e0e\u65b0\u683c\u5f0f\u4f1a\u540c\u65f6\u5b58\u5728\uff0c\u591a\u8f6e\u5b9e\u9a8c\u8fd8\u4f1a\u4ea7\u751f\u591a\u4e2a\u68c0\u67e5\u70b9\u548c\u9884\u6d4b\u7248\u672c\uff0c\u56e0\u6b64\u5de5\u4f5c\u76d8\u5fc5\u987b\u9884\u7559\u5197\u4f59\uff0c\u5e76\u5b9a\u671f\u6838\u5bf9\u6587\u4ef6\u6821\u9a8c\u548c\u3002\u82e5\u591a\u5f20 GPU \u5171\u7528\u4e00\u4e2a\u4f4e\u901f\u7f51\u7edc\u76d8\uff0c\u589e\u52a0\u663e\u5361\u672a\u5fc5\u52a0\u901f\uff0c\u53cd\u800c\u53ef\u80fd\u8ba9\u6240\u6709\u8fdb\u7a0b\u540c\u65f6\u7b49\u5f85 I\/O\u3002<\/p>\n

\u591a GPU \u4e5f\u4e0d\u662f\u628a\u5355\u5361\u65f6\u95f4\u76f4\u63a5\u9664\u4ee5\u5361\u6570\u3002\u9700\u8981\u5148\u786e\u8ba4\u6570\u636e\u80fd\u5426\u5747\u5300\u5207\u5206\u3001\u6bcf\u4e2a\u8fdb\u7a0b\u662f\u5426\u91cd\u590d\u7f13\u5b58\u3001\u901a\u4fe1\u662f\u5426\u7ecf\u8fc7\u9ad8\u901f\u4e92\u8054\uff0c\u4ee5\u53ca\u5931\u8d25\u540e\u80fd\u5426\u4ece\u6700\u8fd1\u68c0\u67e5\u70b9\u7ee7\u7eed\u3002\u5bf9\u65b0\u9879\u76ee\u66f4\u5b9e\u7528\u7684\u505a\u6cd5\u662f\u5148\u56fa\u5b9a\u4e00\u6279\u4ee3\u8868\u8c31\u56fe\uff0c\u5206\u522b\u7528 1 \u5361\u548c 2 \u5361\u8dd1\u76f8\u540c\u914d\u7f6e\uff1b\u53ea\u6709\u541e\u5410\u3001GPU \u5229\u7528\u7387\u548c\u7ed3\u679c\u4e00\u81f4\u6027\u90fd\u7b26\u5408\u9884\u671f\uff0c\u518d\u6269\u5927\u5230 4 \u5361\u6216\u66f4\u591a\u3002<\/p>\n

\u4ee5\u4e0b\u662f\u9879\u76ee\u6392\u961f\u7684\u8d77\u70b9\uff0c\u4e0d\u662f\u8bba\u6587\u7ed9\u51fa\u7684\u786c\u4ef6\u95e8\u69db\u6216\u901f\u5ea6\u4fdd\u8bc1\uff1a\u9884\u8bad\u7ec3\u63a8\u7406\u53ef\u5148\u7528 1 \u5f20 24\u201348 GB GPU\u300116 CPU\u300164 GB \u5185\u5b58\u548c 1 TB NVMe\uff1b\u591a\u6279\u6b21\u6216\u5fae\u8c03\u53ef\u4ece 2\u20144 \u5f20 GPU\u300132 CPU\u3001128 GB \u5185\u5b58\u8d77\u6d4b\u3002\u82e5\u8981\u5904\u7406 TB \u7ea7 MSDT \u6216\u8bad\u7ec3\u6570\u5343\u4e07 PSM\uff0c\u5e94\u6309\u4e00\u6b21\u5c0f\u6837\u5b9e\u6d4b\u7684 GPU \u5229\u7528\u7387\u3001\u5cf0\u503c\u663e\u5b58\u548c\u8bfb\u5199\u5e26\u5bbd\u518d\u6269\u5bb9\u3002<\/p>\n

\u53ef\u6267\u884c\u7684\u6700\u5c0f\u63a8\u7406\u8def\u5f84<\/h2>\n

\u5b98\u65b9\u4ed3\u5e93\u63d0\u4f9b CUDA 12 \u955c\u50cf\u3001\u6a21\u578b\u548c 1 \u4e07\u5f20\u8c31\u56fe\u6d4b\u8bd5\u6587\u4ef6\u3002\u5148\u8dd1\u901a\u5c0f\u6837\u5e76\u4fdd\u5b58\u6821\u9a8c\u548c\uff0c\u518d\u66ff\u6362\u4e3a\u81ea\u5df1\u7684\u6570\u636e\u3002<\/p>\n

git clone https:\/\/github.com\/guomics-lab\/MassNet-DDA.git\r\ncd MassNet-DDA\r\n\r\ndocker pull guomics2017\/massnet-dda:cuda12_v1.1\r\npython download_data.py --prefer-mirror \\\r\n  --files XuanjiNovo_100M_massnet.ckpt bacillus.10k.mgf\r\nsha256sum XuanjiNovo_100M_massnet.ckpt\r\n\r\npython -m XuanjiNovo.XuanjiNovo \\\r\n  --mode=denovo \\\r\n  --peak_path=.\/bacillus.10k.mgf \\\r\n  --model=.\/XuanjiNovo_100M_massnet.ckpt \\\r\n  --output=.\/demo_output\r\n<\/code><\/pre>\n
\u6b63\u5f0f\u4e0a\u91cf\u524d\u8bb0\u5f55\u56db\u9879\uff1a\u8c31\u56fe\u6570\u3001\u6bcf\u79d2\u5904\u7406\u91cf\u3001\u5cf0\u503c\u663e\u5b58\u3001\u8f93\u5165\u76d8\u8bfb\u53d6\u901f\u5ea6\u3002\u5b98\u65b9\u4ed3\u5e93\u62a5\u544a\u5728 NVIDIA H200 \u4e0a\u7ea6\u201c280 spectra per GPU\u201d\uff0c\u4f46\u6ca1\u6709\u5728\u8fd9\u53e5\u4e2d\u660e\u786e\u65f6\u95f4\u5355\u4f4d\uff0c\u56e0\u6b64\u4e0d\u80fd\u76f4\u63a5\u6362\u7b97\u4f60\u7684\u5de5\u671f\uff1b\u4eea\u5668\u3001\u5cf0\u6570\u3001beam\u3001\u7cbe\u4fee\u6b21\u6570\u548c\u5b58\u50a8\u90fd\u4f1a\u6539\u53d8\u541e\u5410\u3002<\/p>\n
\u9002\u7528\u8fb9\u754c\uff1aAI \u9884\u6d4b\u4e0d\u662f\u6570\u636e\u5e93\u641c\u7d22\u7684\u66ff\u4ee3\u58f0\u660e<\/h2>\n
XuanjiNovo \u7684\u4f18\u52bf\u6765\u81ea\u8bba\u6587\u8bbe\u5b9a\u4e0b\u7684\u57fa\u51c6\uff0c\u4e0d\u4ee3\u8868\u6240\u6709\u4eea\u5668\u548c\u6837\u672c\u90fd\u540c\u6837\u63d0\u5347\u3002\u8de8\u5e73\u53f0\u6570\u636e\u53ef\u80fd\u9700\u8981\u5fae\u8c03\uff1b\u8bad\u7ec3\u96c6\u4e0e\u6d4b\u8bd5\u96c6\u5fc5\u987b\u6309\u80bd\u6bb5\u5e8f\u5217\u9694\u79bb\uff0c\u907f\u514d\u770b\u4f3c\u9ad8\u5206\u7684\u6cc4\u6f0f\u3002de novo \u9884\u6d4b\u8fd8\u5e94\u7ed3\u5408\u8d28\u91cf\u8bef\u5dee\u3001FDR\u3001\u6570\u636e\u5e93\u641c\u7d22\u6216\u5408\u6210\u80bd\u9a8c\u8bc1\uff0c\u4e0d\u80fd\u628a\u6a21\u578b\u8f93\u51fa\u76f4\u63a5\u5199\u6210\u65b0\u86cb\u767d\u53d1\u73b0\u6216\u673a\u5236\u8bc1\u636e\u3002<\/p>\n
\u628a\u8c31\u56fe\u89c4\u6a21\u5e26\u6765\uff0c\u5148\u5224\u65ad\u54ea\u4e00\u5c42\u503c\u5f97\u6295\u5165<\/h2>\n
\u5982\u679c\u4f60\u51c6\u5907\u90e8\u7f72 XuanjiNovo\u3001\u8f6c\u6362 MSDT\uff0c\u6216\u8bc4\u4f30\u662f\u5426\u503c\u5f97\u5fae\u8c03\uff0c\u53ef\u5148\u63d0\u4f9b\u4eea\u5668\u4e0e\u788e\u88c2\u65b9\u5f0f\u3001\u6587\u4ef6\u683c\u5f0f\u3001\u8c31\u56fe\u6570\u91cf\u3001\u6837\u672c\u6279\u6b21\u3001\u76ee\u6807\u7269\u79cd\u3001\u73b0\u6709 GPU\/\u5b58\u50a8\u548c\u5e0c\u671b\u5f97\u5230\u7684\u7ed3\u679c\uff1b\u9996\u6b21\u6c9f\u901a\u4e0d\u5fc5\u4f20\u5b8c\u6574\u654f\u611f\u539f\u59cb\u6570\u636e\u3002<\/p>\n
\u4e91\u751f\u4fe1\u53ef\u5148\u534f\u52a9\u5224\u65ad\u662f\u9884\u8bad\u7ec3\u63a8\u7406\u3001\u5e73\u53f0\u5fae\u8c03\u8fd8\u662f\u5927\u89c4\u6a21\u8bad\u7ec3\uff0c\u5e76\u636e\u6b64\u62c6\u5206 CUDA \u73af\u5883\u3001GPU\u3001CPU\u3001\u5185\u5b58\u548c NVMe \u5de5\u4f5c\u76d8\u3002\u540e\u7eed\u53ef\u9009\u62e9\u5728\u4e91\u7aef\u670d\u52a1\u5668\u90e8\u7f72\u5e76\u4ea4\u63a5\u73af\u5883\u3001\u79df\u7528\u670d\u52a1\u5668\u81ea\u884c\u8fd0\u884c\uff0c\u6216\u5728\u786e\u8ba4\u6570\u636e\u4e0e\u4ea4\u4ed8\u8303\u56f4\u540e\u59d4\u6258\u5206\u6790\uff0c\u907f\u514d\u4e00\u5f00\u59cb\u5c31\u4e3a\u4e0d\u5fc5\u8981\u7684\u5168\u91cf\u8bad\u7ec3\u4ed8\u51fa\u65f6\u95f4\u3002<\/p>\n
\u8054\u7cfb\u4e91\u751f\u4fe1\uff0c\u54a8\u8be2 AI \u86cb\u767d\u7ec4\u5b66\u7b97\u529b\u4e0e\u5206\u6790\u65b9\u6848<\/a><\/p>\n
\u53c2\u8003\u8d44\u6599\uff1a<\/p>\n
    \n
  1. MassNet: billion-scale AI-ready mass spectral corpus, Nature Methods, 2026<\/a><\/li>\n
  2. MassNet-DDA \u4e0e XuanjiNovo \u5b98\u65b9\u4ed3\u5e93<\/a><\/li>\n
  3. MSDT \u8f6c\u6362\u5de5\u5177\u5b98\u65b9\u4ed3\u5e93<\/a><\/li>\n<\/ol>\n<\/div>\n","protected":false},"excerpt":{"rendered":"
    Nature Methods 2026 MassNet\u6c47\u603b15.4\u4ebf\u5f20MS\/MS\u8c31\u56fe\u548c\u7ea630TB\u539f\u59cb\u6570\u636e\uff0c\u5e76\u75281\u4ebfPSM\u8bad\u7ec3XuanjiNovo\u3002\u672c\u6587\u62c6\u89e3\u63a8\u7406\u3001\u5fae\u8c03\u3001\u8bad\u7ec3\u4e0eGPU\/NVMe\u89c4\u5212\u3002<\/p>\n","protected":false},"author":1,"featured_media":66335,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_feature_clip_id":0,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_post_was_ever_published":false},"categories":[1,75],"tags":[455,456,223,453,458,454,459,457],"class_list":["post-66336","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-part","category-compute-environment","tag-ai","tag-de-novo","tag-gpu-server","tag-massnet","tag-msdt","tag-xuanjinovo","tag-gpu","tag-457"],"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"https:\/\/i0.wp.com\/www.biocloudservice.com\/wordpress\/wp-content\/uploads\/2026\/11\/daily-bio-blog-20261103-massnet-ai-proteomics-massnet-poster.png?fit=1080%2C1350&ssl=1","_links":{"self":[{"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/posts\/66336","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/comments?post=66336"}],"version-history":[{"count":1,"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/posts\/66336\/revisions"}],"predecessor-version":[{"id":66337,"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/posts\/66336\/revisions\/66337"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/media\/66335"}],"wp:attachment":[{"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/media?parent=66336"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/categories?post=66336"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.biocloudservice.com\/wordpress\/wp-json\/wp\/v2\/tags?post=66336"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}