发布求助
文献互助 智能选刊 最新文献

Patterns最新文献

筛选
英文 中文
Generalization of biological foundation models requires prospective validation across contexts. 生物基础模型的推广需要跨环境的前瞻性验证。
IF 10.8
Patterns Pub Date : 2026-10-09 DOI: 10.1016/j.patter.2026.101665
Lucas A Gillenwater, Casey S Greene
{"title":"Generalization of biological foundation models requires prospective validation across contexts.","authors":"Lucas A Gillenwater, Casey S Greene","doi":"10.1016/j.patter.2026.101665","DOIUrl":"10.1016/j.patter.2026.101665","url":null,"abstract":"<p><p>Biological foundation models promise generalization. The authors argue that the field needs a community-driven evaluation that tests generalization through prospective validation across biological and experimental contexts.</p>","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"7 10","pages":"101665"},"PeriodicalIF":10.8,"publicationDate":"2026-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674296/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Ethical principles and considerations in health AI. 卫生人工智能中的伦理原则和考虑。
IF 10.8
Patterns Pub Date : 2026-10-09 DOI: 10.1016/j.patter.2026.101662
Yilin Ning, Yian Ma, Mingxuan Liu, Gary S Collins, Arun James Thirunavukarasu, Melissa D McCradden, Jasmine Chiat Ling Ong, Yuhe Ke, Daniel Shu Wei Ting, Danielle S Bitterman, Yih Chung Tham, Alastair K Denniston, Pearse A Keane, Josip Car, Patrick Tan, Karel G M Moons, Tien Yin Wong, Eric J Topol, Nan Liu
{"title":"Ethical principles and considerations in health AI.","authors":"Yilin Ning, Yian Ma, Mingxuan Liu, Gary S Collins, Arun James Thirunavukarasu, Melissa D McCradden, Jasmine Chiat Ling Ong, Yuhe Ke, Daniel Shu Wei Ting, Danielle S Bitterman, Yih Chung Tham, Alastair K Denniston, Pearse A Keane, Josip Car, Patrick Tan, Karel G M Moons, Tien Yin Wong, Eric J Topol, Nan Liu","doi":"10.1016/j.patter.2026.101662","DOIUrl":"10.1016/j.patter.2026.101662","url":null,"abstract":"<p><p>Ethical principles are foundations of medicine and should guide the integration and implementation of health AI but are often considered in isolation, typically at deployment. The authors argue that ethical principles are fundamental constructs across the full AI development-deployment life cycle and show how trade-offs and dependencies between principles shape feasible and ethically robust development and use of health AI, particularly in low-resource settings.</p>","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"7 10","pages":"101662"},"PeriodicalIF":10.8,"publicationDate":"2026-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674299/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217025","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Erratum: DestinyNet: A deep-learning framework for cell-fate analysis from lineage-tracing single-cell RNA sequencing data. 勘误：DestinyNet：从谱系追踪单细胞RNA测序数据进行细胞命运分析的深度学习框架。
IF 10.8
Patterns Pub Date : 2026-09-29 eCollection Date: 2026-10-09 DOI: 10.1016/j.patter.2026.101692
Zuozhu Liu, Songtao Jiang, Tianxiang Hu, Ziwei Xue, Yuhui Chen, Dengfeng Ruan, Haoxiang Xia, Yang Feng, Jin Hao, Wei Li, Joey Tianyi Zhou, Bing Fang, Jian Wu, Wanlu Liu
{"title":"Erratum: DestinyNet: A deep-learning framework for cell-fate analysis from lineage-tracing single-cell RNA sequencing data.","authors":"Zuozhu Liu, Songtao Jiang, Tianxiang Hu, Ziwei Xue, Yuhui Chen, Dengfeng Ruan, Haoxiang Xia, Yang Feng, Jin Hao, Wei Li, Joey Tianyi Zhou, Bing Fang, Jian Wu, Wanlu Liu","doi":"10.1016/j.patter.2026.101692","DOIUrl":"https://doi.org/10.1016/j.patter.2026.101692","url":null,"abstract":"<p><p>[This corrects the article DOI: 10.1016/j.patter.2025.101471.].</p>","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"7 10","pages":"101692"},"PeriodicalIF":10.8,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674297/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Unifying AI-assisted scientific discovery around exploration, hypothesis generation, and testing. 围绕探索、假设生成和测试，统一人工智能辅助的科学发现。
IF 10.8
Patterns Pub Date : 2026-09-28 eCollection Date: 2026-10-09 DOI: 10.1016/j.patter.2026.101660
Md Musaddaqul Hasib, Sumin Jo, Harsh Sinha, Jifeng Song, Arun Das, Zhentao Liu, Hugh Galloway, Huey Huang, Jianqiu Michelle Zhang, Kexun Zhang, Shou-Jiang Gao, Yu-Chiao Chiu, Lei Li, Yufei Huang
{"title":"Unifying AI-assisted scientific discovery around exploration, hypothesis generation, and testing.","authors":"Md Musaddaqul Hasib, Sumin Jo, Harsh Sinha, Jifeng Song, Arun Das, Zhentao Liu, Hugh Galloway, Huey Huang, Jianqiu Michelle Zhang, Kexun Zhang, Shou-Jiang Gao, Yu-Chiao Chiu, Lei Li, Yufei Huang","doi":"10.1016/j.patter.2026.101660","DOIUrl":"10.1016/j.patter.2026.101660","url":null,"abstract":"<p><p>Large language models (LLMs) and agent systems increasingly support scientific discovery across domains. Yet the literature remains fragmented around isolated tasks, with limited attention to how components form integrated workflows. Here, we introduce exploration, hypothesis generation, and testing (EXHYTE) as an empirical workflow abstraction for reviewing AI-assisted scientific discovery. We mapped a corpus of recent studies to EXHYTE stages and substages to identify recurring strategies, workflow connections, and capability gaps. The analysis shows strong progress in exploration, especially retrieval, knowledge assembly, and representation learning as well as growing capacity for hypothesis and idea generation. The central bottleneck remains the transition from hypothesis or idea generation to executable experimental or computational testing, followed by feedback that can refine subsequent workflow steps. Because EXHYTE organizes systems by workflow function rather than agent architecture, it applies to both multi-agent scaffolds and emerging generalist tool-using agents. It also shows that integrated workflows do not imply full automation: current systems still depend on human judgment for problem formulation, contextualization, interpretation, risk assessment, and oversight. We discuss evaluation, reproducibility, hypothesis-space narrowing, external validation, and governance considerations.</p>","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"7 10","pages":"101660"},"PeriodicalIF":10.8,"publicationDate":"2026-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674306/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149216958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
AI safety is a system property, not a sum of individual defenses. 人工智能安全是一个系统属性，而不是个体防御的总和。
IF 10.8
Patterns Pub Date : 2026-09-25 eCollection Date: 2026-10-09 DOI: 10.1016/j.patter.2026.101686
Xiang Chen, Chenxi Lin, Longlong Zhu, Yiheng Xu, Jiajie Su, Chunming Wu
{"title":"AI safety is a system property, not a sum of individual defenses.","authors":"Xiang Chen, Chenxi Lin, Longlong Zhu, Yiheng Xu, Jiajie Su, Chunming Wu","doi":"10.1016/j.patter.2026.101686","DOIUrl":"10.1016/j.patter.2026.101686","url":null,"abstract":"<p><p>Safety for large language models (LLMs) and artificial intelligence (AI) agents remains fragmented; defenses against jailbreaking coexist with unresolved hallucination, fairness, and autonomous-control problems. Existing frameworks address these risks, but few specify how failure in one dimension changes another within a deployment. We argue that distinct safety failures can interact and amplify each other. We identify four safety challenge classes, analyze their interactions, and ground the argument in real-world incidents involving compound failures. We then audit GPT-5 and Claude-Sonnet-4.6 on clinical vignettes from the Medical Question Answering benchmark based on the United States Medical Licensing Examination (MedQA-USMLE). A single jailbreak can affect multiple safety dimensions, although patterns vary by model and attack. We propose three design principles for integrated safety architectures and identify agent safety and mechanistic interpretability as urgent open frontiers.</p>","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"7 10","pages":"101686"},"PeriodicalIF":10.8,"publicationDate":"2026-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674307/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149216936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Don't black box your research. 不要把你的研究黑箱化。
IF 10.8
Patterns Pub Date : 2026-09-11 DOI: 10.1016/j.patter.2026.101663
Alejandra Alvarado, Andrew L Hufton
{"title":"Don't black box your research.","authors":"Alejandra Alvarado, Andrew L Hufton","doi":"10.1016/j.patter.2026.101663","DOIUrl":"https://doi.org/10.1016/j.patter.2026.101663","url":null,"abstract":"","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"7 9","pages":"101663"},"PeriodicalIF":10.8,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13576648/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148978239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
What-if effects: A tale of entangled covariates. 假设效应：纠缠协变量的故事。
IF 10.8
Patterns Pub Date : 2026-09-11 DOI: 10.1016/j.patter.2026.101616
Zhaowei Yu, Yong Zhang
{"title":"What-if effects: A tale of entangled covariates.","authors":"Zhaowei Yu, Yong Zhang","doi":"10.1016/j.patter.2026.101616","DOIUrl":"10.1016/j.patter.2026.101616","url":null,"abstract":"<p><p>Single-cell epigenomic studies promise to reveal how disease states, donor traits, environmental exposures, and technical factors shape cellular regulatory programs, yet these variables are often highly collinear and difficult to disentangle, particularly in human cohorts. The recent study by Møller and Madsen in <i>Patterns</i> introduces DeepDive, a probabilistic deep-learning framework that disentangles known covariate effects and residual variations in single-nucleus ATAC-seq data, enabling counterfactual \"what-if\" analyses of chromatin accessibility.</p>","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"7 9","pages":"101616"},"PeriodicalIF":10.8,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13576651/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148978324","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Erratum: RDMkit: A research data management toolkit for life sciences. 勘误：RDMkit：生命科学研究数据管理工具包。
IF 10.8
Patterns Pub Date : 2026-09-09 eCollection Date: 2026-10-09 DOI: 10.1016/j.patter.2026.101687
Pinar Alper, Flora D'Anna, Bert Droesbeke, Munazah Andrabi, Rafael Andrade Buono, Federico Bianchini, Korbinian Bösl, Ishwar Chandramouliswaran, Martin Cook, Daniel Faria, Nazeefa Fatima, Rob Hooft, Niclas Jareborg, Mijke Jetten, Diana Pilvar, Gil Poiares-Oliveira, Marina Popleteeva, Laura Portell-Silva, Jan Slifka, Marek Suchánek, Celia van Gelder, Danielle Welter, Ulrike Wittig, Frederik Coppens, Carole Goble
{"title":"Erratum: RDMkit: A research data management toolkit for life sciences.","authors":"Pinar Alper, Flora D'Anna, Bert Droesbeke, Munazah Andrabi, Rafael Andrade Buono, Federico Bianchini, Korbinian Bösl, Ishwar Chandramouliswaran, Martin Cook, Daniel Faria, Nazeefa Fatima, Rob Hooft, Niclas Jareborg, Mijke Jetten, Diana Pilvar, Gil Poiares-Oliveira, Marina Popleteeva, Laura Portell-Silva, Jan Slifka, Marek Suchánek, Celia van Gelder, Danielle Welter, Ulrike Wittig, Frederik Coppens, Carole Goble","doi":"10.1016/j.patter.2026.101687","DOIUrl":"https://doi.org/10.1016/j.patter.2026.101687","url":null,"abstract":"<p><p>[This corrects the article DOI: 10.1016/j.patter.2025.101345.].</p>","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"7 10","pages":"101687"},"PeriodicalIF":10.8,"publicationDate":"2026-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674302/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Designing reproducible large-language-model-assisted scientific analyses. 设计可重复的大语言模型辅助科学分析。
IF 10.8
Patterns Pub Date : 2026-09-04 eCollection Date: 2026-09-11 DOI: 10.1016/j.patter.2026.101644
Casey W Dunn, Darrin T Schultz, Jacob M Musser
{"title":"Designing reproducible large-language-model-assisted scientific analyses.","authors":"Casey W Dunn, Darrin T Schultz, Jacob M Musser","doi":"10.1016/j.patter.2026.101644","DOIUrl":"10.1016/j.patter.2026.101644","url":null,"abstract":"<p><p>Large language models (LLMs) increasingly write code, analyze data, and orchestrate scientific workflows. This can create reproducibility challenges because LLMs blur the boundary between how an analysis is built and what it depends on at run time. Guidelines for LLM-assisted science begin with a critical choice: whether the LLM sits <i>on</i> the data path of the published analysis, a live step results depend on, or <i>off</i> the data path, producing durable artifacts such as code. Reproducible analyses require preserving data, code, and runtime; an on-path LLM becomes part of the runtime, a dependency that may change, be deprecated, or become inaccessible. We derive six recommendations: (1) keep the LLM off the data path where possible, (2) preserve LLM-generated artifacts, (3) verify results by methods suited to the LLM's role, (4) consider open-weight models, (5) record the model version, and (6) assess determinism.</p>","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"7 9","pages":"101644"},"PeriodicalIF":10.8,"publicationDate":"2026-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13576654/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148978238","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Identifying low-quality features and suboptimal models to enhance phenotype prediction. 识别低质量特征和次优模型以增强表型预测。
IF 10.8
Patterns Pub Date : 2026-09-03 eCollection Date: 2026-10-09 DOI: 10.1016/j.patter.2026.101643
Baining Shen, Min Li, Xiaoyue Hu, Ming Chen, Wei-Hua Chen, Hangjin Jiang
{"title":"Identifying low-quality features and suboptimal models to enhance phenotype prediction.","authors":"Baining Shen, Min Li, Xiaoyue Hu, Ming Chen, Wei-Hua Chen, Hangjin Jiang","doi":"10.1016/j.patter.2026.101643","DOIUrl":"10.1016/j.patter.2026.101643","url":null,"abstract":"<p><p>It is common to build predictive models for complex phenotypes, such as diseases, where sample sizes are often limited by the cost of data collection. Machine learning and statistical methods are widely used in these cases. However, they may fail to identify high-quality features and lead to suboptimal models with high underfitting or overfitting risks. Here, we present mBoost to identify low-quality features and suboptimal models to enhance phenotype prediction. Importantly, it provides confidence on model performance, helping researchers avoid the frustrating cycle of model switching for possibly better performance. We demonstrate mBoost with simulations and real-data examples. Specifically, mBoost identifies optimal depth and width for a deep neural network in a drug example, detects suboptimal models in microbiome cohorts, and proposes a measure for telling whether the trained model is applicable to new data. Thus, mBoost will serve as a practical tool for model building and diagnosis.</p>","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"7 10","pages":"101643"},"PeriodicalIF":10.8,"publicationDate":"2026-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674310/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149216966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
下一页 尾页
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整，为了账户安全，请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们：info@booksci.cn Book学术提供免费学术资源搜索服务，方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号：604180095
Book学术官方微信
小红书