Nature报道：谷歌新模型1秒读懂DNA变异！首次统一基因组全任务

市场资讯2025-06-27

炒股就看金麒麟分析师研报，权威，专业，及时，全面，助您挖掘潜力主题机会！谷歌DeepMind Alpha家族又双叒登上Nature报道，这次瞄准的是DNA变异。现在只需1秒，就能精确定位基因组序列变异。据Nature最新报道，谷歌DeepMind团队目前推出了突破性生物模型AlphaGenome。它能够从长达1兆碱基的DNA序列中，同时预测数千种功能基因组特征，并以单碱基分辨率评估变异效应。在...

网页链接

免责声明：本文观点仅代表作者个人观点，不构成本平台的投资建议，本平台不对文章信息准确性、完整性和及时性做出任何保证，亦不对因使用或信赖文章信息引发的任何损失承担责任。

精彩评论

我们需要你的真知灼见来填补这片空白

发表看法

{"i18n":{"language":"zh_CN"},"isChannel":false,"data":{"share":"https://www.laohu8.com/m/news/2546715452?lang=zh_CN&edition=full","thumbnail":"","is_english":false,"pubTime":"2025-06-27 15:32","share_image_url":"https://static.laohu8.com/b0d1b7e8843deea78cc308b15114de44","id":"2546715452","market":"fut","top_or_hot":-1,"title":"Nature报道：谷歌新模型1秒读懂DNA变异！首次统一基因组全任务","media":"市场资讯","content":"<div>\n<p>炒股就看金麒麟分析师研报，权威，专业，及时，全面，助您挖掘潜力主题机会！ 谷歌DeepMind Alpha家族又双叒登上Nature报道，这次瞄准的是DNA变异。现在只需1秒，就能精确定位基因组序列变异。据Nature最新报道，谷歌DeepMind团队目前推出了突破性生物模型AlphaGenome。它能够从长达1兆碱基的DNA序列中，同时预测数千种功能基因组特征，并以单碱基分辨率评估变异效应。在...</p>\n\n<a href=\"https://finance.sina.com.cn/stock/t/2025-06-27/doc-infcpark0580462.shtml\">网页链接</a>\n\n</div>\n","source":"sina","html":"<!DOCTYPE html>\n<html>\n<head>\n<meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<meta name=\"viewport\" content=\"width=device-width,initial-scale=1.0,minimum-scale=1.0,maximum-scale=1.0,user-scalable=no\"/>\n<meta name=\"format-detection\" content=\"telephone=no,email=no,address=no\" />\n<title>Nature报道：谷歌新模型1秒读懂DNA变异！首次统一基因组全任务</title>\n<style type=\"text/css\">\na,abbr,acronym,address,applet,article,aside,audio,b,big,blockquote,body,canvas,caption,center,cite,code,dd,del,details,dfn,div,dl,dt,\nem,embed,fieldset,figcaption,figure,footer,form,h1,h2,h3,h4,h5,h6,header,hgroup,html,i,iframe,img,ins,kbd,label,legend,li,mark,menu,nav,\nobject,ol,output,p,pre,q,ruby,s,samp,section,small,span,strike,strong,sub,summary,sup,table,tbody,td,tfoot,th,thead,time,tr,tt,u,ul,var,video{ font:inherit;margin:0;padding:0;vertical-align:baseline;border:0 }\nbody{ font-size:16px; line-height:1.5; color:#999; background:transparent; }\n.wrapper{ overflow:hidden;word-break:break-all;padding:10px; }\nh1,h2{ font-weight:normal; line-height:1.35; margin-bottom:.6em; }\nh3,h4,h5,h6{ line-height:1.35; margin-bottom:1em; }\nh1{ font-size:24px; }\nh2{ font-size:20px; }\nh3{ font-size:18px; }\nh4{ font-size:16px; }\nh5{ font-size:14px; }\nh6{ font-size:12px; }\np,ul,ol,blockquote,dl,table{ margin:1.2em 0; }\nul,ol{ margin-left:2em; }\nul{ list-style:disc; }\nol{ list-style:decimal; }\nli,li p{ margin:10px 0;}\nimg{ max-width:100%;display:block;margin:0 auto 1em; }\nblockquote{ color:#B5B2B1; border-left:3px solid #aaa; padding:1em; }\nstrong,b{font-weight:bold;}\nem,i{font-style:italic;}\ntable{ width:100%;border-collapse:collapse;border-spacing:1px;margin:1em 0;font-size:.9em; }\nth,td{ padding:5px;text-align:left;border:1px solid #aaa; }\nth{ font-weight:bold;background:#5d5d5d; }\n.symbol-link{font-weight:bold;}\n/* header{ border-bottom:1px solid #494756; } */\n.title{ margin:0 0 8px;line-height:1.3;color:#ddd; }\n.meta {color:#5e5c6d;font-size:13px;margin:0 0 .5em; }\na{text-decoration:none; color:#2a4b87;}\n.meta .head { display: inline-block; overflow: hidden}\n.head .h-thumb { width: 30px; height: 30px; margin: 0; padding: 0; border-radius: 50%; float: left;}\n.head .h-content { margin: 0; padding: 0 0 0 9px; float: left;}\n.head .h-name {font-size: 13px; color: #eee; margin: 0;}\n.head .h-time {font-size: 11px; color: #7E829C; margin: 0;line-height: 11px;}\n.small {font-size: 12.5px; display: inline-block; transform: scale(0.9); -webkit-transform: scale(0.9); transform-origin: left; -webkit-transform-origin: left;}\n.smaller {font-size: 12.5px; display: inline-block; transform: scale(0.8); -webkit-transform: scale(0.8); transform-origin: left; -webkit-transform-origin: left;}\n.bt-text {font-size: 12px;margin: 1.5em 0 0 0}\n.bt-text p {margin: 0}\n</style>\n</head>\n<body>\n<div class=\"wrapper\">\n<header>\n<h2 class=\"title\">\nNature报道：谷歌新模型1秒读懂DNA变异！首次统一基因组全任务\n</h2>\n\n<h4 class=\"meta\">\n\n\n2025-06-27 15:32 北京时间&nbsp;&nbsp;&nbsp;<a href=https://finance.sina.com.cn/stock/t/2025-06-27/doc-infcpark0580462.shtml><strong>市场资讯</strong></a>\n\n\n</h4>\n\n</header>\n<article>\n<div>\n<p>炒股就看金麒麟分析师研报，权威，专业，及时，全面，助您挖掘潜力主题机会！ 谷歌DeepMind Alpha家族又双叒登上Nature报道，这次瞄准的是DNA变异。现在只需1秒，就能精确定位基因组序列变异。据Nature最新报道，谷歌DeepMind团队目前推出了突破性生物模型AlphaGenome。它能够从长达1兆碱基的DNA序列中，同时预测数千种功能基因组特征，并以单碱基分辨率评估变异效应。在...</p>\n\n<a href=\"https://finance.sina.com.cn/stock/t/2025-06-27/doc-infcpark0580462.shtml\">网页链接</a>\n\n</div>\n\n\n</article>\n</div>\n</body>\n</html>\n","isBrief":false,"type":0,"news_type":1,"symbol":"LU1066051498.USD","symbol_name":"HSBC GIF GLOBAL EQUITY VOLATILITY FOCUSED \"AM2\" (USD) INC","start_time":0,"source_url":"https://finance.sina.com.cn/stock/t/2025-06-27/doc-infcpark0580462.shtml","article_id":"2546715452","we_media_id":null,"thumbnails":[],"rights":null,"url":"https://stock-news.laohu8.com/highlight/detail?id=2546715452","pubTimestamp":1751009520,"columns":[],"sourceInfo":{"source_id":"sina","name":"sina"},"weMediaInfo":null,"summary":"谷歌DeepMind Alpha家族又双叒登上Nature报道，这次瞄准的是DNA变异。据Nature最新报道，谷歌DeepMind团队目前推出了突破性生物模型AlphaGenome。它能够从长达1兆碱基的DNA序列中，同时预测数千种功能基因组特征，并以单碱基分辨率评估变异效应。结果表明，AlphaGenome在多模态和专门的单模态任务上都存在优势，可以准确模拟基因组轨道和变异效应。基于AlphaGenome的多维度剪接预测，实验团队为每个预测模态设计了定制的变异评分策略，并将单个评分求和，以综合考量变异预测效应。","collect":0,"end_time":0,"defaultTopTitle":"sina.com.cn","property":[],"viewcount":null,"language":"zh","relate_stocks":{"LU1066051498.USD":"HSBC GIF GLOBAL EQUITY VOLATILITY FOCUSED \"AM2\" (USD) INC","LU1196500208.SGD":"NORDEA STABLE RETURN \"HB\" (SGDHDG) ACC","LU0965508806.USD":"AB LOW VOLATILITY EQUITY PORTFOLIO \"AD\" (USD) INC","LU1066053197.SGD":"HSBC GIF GLOBAL EQUITY VOLATILITY FOCUSED \"AM3\" (SGDHDG) INC","BK4077":"互动媒体与服务","GOOG":"谷歌","LU0456855351.SGD":"JPMorgan Funds - Global Equity A (acc) SGD","LU1059921491.USD":"NORDEA 1 GLOBAL STABLE EQUITY \"HB\" (USDHDG) ACC","LU0323591593.USD":"SCHRODER ISF QEP GLOBAL QUALITY \"A\" (USD) ACC","LU0198837287.USD":"UBS (LUX) EQUITY SICAV - USA GROWTH \"P\" (USD) ACC","GOOGL":"谷歌A","BK4514":"搜索引擎","LU0985320562.USD":"NORDEA 1 GLOBAL STARS EQUITY \"BP\" (USD) ACC","LU1116320737.USD":"BGF SYSTEMATIC GLOBAL ENHANCED EQUITY YIELD \"A6\" (USD) INC","LU0265550946.USD":"BGF SYSTEMATIC GLOBAL ENHANCED EQUITY YIELD \"A5\" (USD) INC","IE00BJJMRX11.SGD":"Janus Henderson Balanced A Acc SGD","IE0004091025.USD":"BNY MELLON GLOBAL OPPORTUNITIES \"B\" (USD) ACC","LU0417517546.SGD":"Allianz US Equity Cl AT Acc SGD","LU2242650005.HKD":"FIDELITY FUNDS GLOBAL MULTI ASSET DYNAMIC \"A\" (HKD) ACC","IE0034235303.USD":"PINEBRIDGE US RESEARCH ENHANCED CORE EQUITY \"A\" (USD) ACC","LU1066051811.HKD":"HSBC GIF GLOBAL EQUITY VOLATILITY FOCUSED \"AM2\" (HKD) INC","LU2360032135.SGD":"ALLSPRING GLOBAL EQUITY ENHANCED INCOME \"A\" (SGDHDG) INC","LU1127390331.HKD":"AB SICAV I - ALL MARKET INCOME PORTFOLIO \"A\" (HKD) ACC","USJW.SI":"ALPHAB 3xLongSG261006","LU0079474960.USD":"联博美国增长基金A","LU0256863811.USD":"ALLIANZ US EQUITY \"A\" INC","LU2054465674.USD":"UBS (LUX) KEY SELEC SICAV DIGITAL TRANSFORMATION T \"P\" (USD) ACC","BK4592":"伊斯兰概念","LU0742534661.SGD":"Fidelity America A-SGD (hedged)","LU0964807845.USD":"ALLIANZ INCOME & GROWTH \"A\" (USD) INC","LU2403377893.USD":"ALLIANZ SELECT INCOME AND GROWTH \"AM\" (USD) INC","IE00B5TLWC47.USD":"BNY MELLON LONG-TERM GLOBAL EQUITY \"B\" (USD) ACC","BK4579":"人工智能","LU0107464264.USD":"abrdn SICAV I - GLOBAL INNOVATION EQUITY \"A\" (USD) ACC","LU1116320901.HKD":"BGF SYSTEMATIC GLOBAL ENHANCED EQUITY YIELD \"A6\" (HKD) INC","LU1221951046.USD":"NORDEA 1 STABLE RETURN \"HM\" (USDHDG) INC","LU1032955483.USD":"NORDEA 1 STABLE RETURN \"HB\" (USDHDG) ACC","LU2125154935.USD":"ALLSPRING (LUX) WF GLOBAL EQUITY ENHANCED INCOME \"I\" (USD) INC","LU1221951129.SGD":"NORDEA 1 STABLE RETURN \"HM\" (SGDHDG) INC","BK4574":"无人驾驶","IE00B775H168.HKD":"JANUS HENDERSON BALANCED \"A5M\" (HKD) INC","LU2063271972.USD":"富兰克林创新领域基金","LU1988902786.USD":"FULLERTON LUX FUNDS GLOBAL ABSOLUTE ALPHA \"I\" (USD) ACC","DNA":"Ginkgo Bioworks Holdings Inc.","LU0052756011.USD":"TEMPLETON GLOBAL BALANCED \"A\" (USD) INC","IE00BJTD4V19.USD":"NEUBERGER BERMAN US LONG SHORT EQUITY \"A1\" (USD) ACC","IE0004445015.USD":"JANUS HENDERSON BALANCED \"A2\" (USD) ACC","BK4573":"虚拟现实","LU0310800965.SGD":"FTIF - Templeton Global Balanced A Acc SGD","BK4581":"高盛持仓"},"translate_title":"Nature reports: Google's new model can read DNA mutations in 1 second! The first unified genome task","themeId":null,"isJumpTheme":false,"ttsUrl":null,"symbols_score_info":{"USJW.SI":0.6,"DNA":1,"GOOG":1,"GOOGL":1},"content_text":"炒股就看金麒麟分析师研报，权威，专业，及时，全面，助您挖掘潜力主题机会！ 谷歌DeepMind Alpha家族又双叒登上Nature报道，这次瞄准的是DNA变异。现在只需1秒，就能精确定位基因组序列变异。据Nature最新报道，谷歌DeepMind团队目前推出了突破性生物模型AlphaGenome。它能够从长达1兆碱基的DNA序列中，同时预测数千种功能基因组特征，并以单碱基分辨率评估变异效应。在基因表达、剪接、染色质可及性等多种任务上性能全面超越现有模型，为解析基因组调控代码提供了强大工具。作者将其描述为整个生物领域的里程碑：在未来，AlphaGenome也会更好地帮助我们理解疾病，癌症这本“天书”也许终于得以破解。解读基因组序列变异的影响始终是生物学领域的核心挑战。过去十年里，科学家们为了揭开基因组的秘密，单独开发了数十种AI模型，那么是否能创造一种“一体化”解释工具呢？AlphaGenome就是这样一个模型，可以将多模态预测、长序列背景和碱基对分辨率统一于单一框架。模型架构受U-Net启发，可以将1兆碱基的DNA输入序列，在下采样阶段处理为两种类型的序列表达，分别是对应线性基因组的一维嵌入（1bp和128bp分辨率），和对应基因组片段空间相互作用的二维嵌入（2048bp分辨率）。在架构内部，卷积层对局部序列模式进行建模，Transformer块则结合Rotary位置编码，对颗粒度较粗但范围更长的依赖关系进行建模。通过8个互相连接的张量处理单元，实现对完整碱基对的分辨率训练，然后利用编码器跳跃连接，在上采样阶段恢复序列的1bp分辨率。最后输出包括基因表达、详细剪接模式、染色质状态和染色质接触图谱在内的11种模态，涵盖5930条人类或1128条小鼠基因组轨道。模型通过预训练和蒸馏两阶段进行训练：最终实现在NVIDIA H100 GPU上，学生模型的推理时间能达到一秒以内，具有极高的效率。为了评估AlphaGenome的泛化能力，研究团队进行了24项基因组轨道评估，将AlphaGenome模型与各任务最强的现有模型进行比较。AlphaGenome在其中22项都保持领先，其中与另一种多模态序列模型Borzoi3相比，更是在细胞类型特异性的LFC预测上，表现出+17.4%的相对改进。在预测变异效应方面，实验组组装了26个变异效应预测基准，包含基因表达、剪接、多聚腺苷酸化、增强子-基因连接、DNA可及性和转录因子结合等。与现有最强模型对比，有24项达到或超越，例如在表达QTL的方向预测，相比Borzoi3提升25.5%，在可及性QTL上相比ChromBPNet10提升8%。结果表明，AlphaGenome在多模态和专门的单模态任务上都存在优势，可以准确模拟基因组轨道和变异效应。另外，AlphaGenome也在跨模态基因组轨道预测方面，都达到了最先进水平。预训练的折叠特异性模型显示，在未见的基因组区间上，预测的读取覆盖度与观察到的读取覆盖度高度一致。从定量角度，人类和小鼠基因组中功能性基因组轨道的预测信号与观测信号之间，存在较强的皮尔逊相关系数（r），整体表达水平预测良好。在剪接模态方面，AlphaGenome首次实现剪接位点、剪接效率和剪接连接的全方位预测，展现出强大的预测组织特异性可变剪接的能力。基于AlphaGenome的多维度剪接预测，实验团队为每个预测模态设计了定制的变异评分策略，并将单个评分求和，以综合考量变异预测效应。在剪接相关变异效应预测（VEP）任务上进行基准测试，AlphaGenome在精细定位的剪接QTL（sQTL）分类中表现最佳，并在监督和无监督场景下均取得最高性能。但在MFASS评估罕见变异是否破坏剪接能力的实验中，AlphaGenome的auPRC达0.54，表现仅略低于Pangolin的0.51，但超过了SpliceAI和DeltaSplice（均为0.49）。总之，AlphaGenome在7项基准测试中的6项上实现了最先进的剪接变异效应预测，为剪接事件改变和转录本结构提供了更全面的视图。AlphaGenome在临床上，可以帮助研究人员更精准地理解疾病的潜在原因，甚至发现新的治疗靶点。例如在一项针对T细胞急性淋巴细胞白血病(T-ALL)的研究中，AlphaGenome就通过引入MYB DNA结合基序，成功解析了TAL1基因附近的致癌变异。另外它还可以帮助预测合成DNA的设计以及协助进行基本DNA研究，未来通过扩展数据，AlphaGenome也将会产生更准确的预测精度以及涵盖更广泛的物种，科学家们将只需要进行微调，就能更快地生成和测试假设。目前AlphaGenome已提供预览版，并计划正式发布，欢迎大家抢先体验。链接：https://macro.com/app/pdf/56a50ffc-120d-4a9a-87e5-a18753430f22代码链接：https://github.com/google-deepmind/alphagenome参考链接：[1]https://x.com/GoogleAI/status/1937895472305152387[2]https://deepmind.google/discover/blog/alphagenome-ai-for-better-understanding-the-genome/[3]https://www.nature.com/articles/d41586-025-01998-w[4]https://www.science.org/doi/10.1126/science.1259037\n\n\n海量资讯、精准解读，尽在新浪财经APP","kind":"news","is_publish_news":true,"is_publish_highlight":false,"is_publish_live":false,"is_publish_wemedia":null,"editions":null,"column":"","sentiment":"0","news_tag":"","news_rank":0,"isVideo":false,"video":null,"symbols":[],"gpt_button":0,"need_auth":false,"need_login_tip":false,"code":"91000000","status":"200"},"commentList":[],"isCommentEnd":true,"newsSizeData":{"likeSize":0,"commentSize":0,"repostSize":0,"favoriteSize":0,"likeStatus":false,"favoriteStatus":false},"APP":{"userAgent":"Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; ClaudeBot/1.0; +claudebot@anthropic.com)","isDev":false,"isTTM":false,"tenantId":"TBCN","deviceId":"web-server-community-laohu8-v3","version":"4.43.1","shortVersion":"4.43.1","platform":"web","vendor":"web","appName":"laohu8","isIOS":false,"isAndroid":false,"isTiger":false,"isTHS":false,"isWeiXin":false,"isWeiXinMini":false,"isWeiBo":false,"isQQ":false,"isBaiduSwan":false,"isBaiduBox":false,"isDingTalk":false,"isToutiao":false,"isOnePlus":false,"isHuaWei":false,"isXiaomi":false,"isXiaomiWebView":false,"isOppo":false,"isVivo":false,"isSamsung":false,"isMobile":false},"href":"/m/news/2546715452","isCrawlerRequest":true}