<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>BioAI 日知 · 虚拟生命科学</title><link>https://bioai-rizhi.pages.dev</link><description>来源原文索引；自动归类不等于事实复核。预印本保留状态。</description><language>zh-cn</language><ttl>60</ttl><atom:link href="https://bioai-rizhi.pages.dev/feed.xml" rel="self" type="application/rss+xml"/><item><guid isPermaLink="false">bioai:0dabf981-c38a-4a97-afe1-7aebd21c2e68</guid><title>Sampling protein language models for functional protein design</title><link>https://doi.org/10.1016/j.cels.2026.101714</link><description>[Crossref REST API] Sampling protein language models for functional protein design</description></item><item><guid isPermaLink="false">bioai:4812035d-bce5-4440-a5ac-b9a78ec1e8db</guid><title>从AI预测到干湿闭环：AI for Science 正在跨越临界点</title><link>https://www.xtalpi.com/%e4%bb%8eai%e9%a2%84%e6%b5%8b%e5%88%b0%e5%b9%b2%e6%b9%bf%e9%97%ad%e7%8e%af%ef%bc%9aai-for-science-%e6%ad%a3%e5%9c%a8%e8%b7%a8%e8%b6%8a%e4%b8%b4%e7%95%8c%e7%82%b9/</link><description>[XtalPi official Chinese RSS] 科学哲学家卡尔·波普尔认为， 科学在猜想与反驳中向前发展 。提出一个新想法只是开始，接下来还得看它能否经得住检验。 如今，AI 让提出假设变得越来越快。候选药物分子、新材料结构，模型可以在短时间内给出大量方案。但在实验室里，确认它们是否有效，仍可能需要数周甚至数月。 AI 正在快速重构科研的上半场，验证能力渐渐成了卡住科研进度的新瓶颈 。这道难题，该如何解答？ 数字世界假设正在变得空前便利 MIT FutureTech 与 Google、Google DeepMind 联合研究团队发布的《AI in Science: Early Insights》，观察</description><pubDate>Wed, 07 Oct 2026 03:30:18 GMT</pubDate></item><item><guid isPermaLink="false">bioai:a2616d91-7385-45ad-b86f-ffdcd6498fd9</guid><title>Isomorphic Labs joins the Virtual Biology Initiative to build foundational data for AI models to predict and treat disease</title><link>https://www.isomorphiclabs.com/articles/isomorphic-labs-joins-the-virtual-biology-initiative</link><description>[Isomorphic Labs official announcements] Listen: Spotify Apple Podcasts Convened by Biohub, participating organisations are investing funding, data, computation and new measurement technology - the largest coordinated commitment to generating AI-ready biological data to date. The result will be an open resource for the </description><pubDate>Wed, 07 Oct 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:122965e1-9134-58c7-8906-3ae840d31811</guid><title>活性提升超千倍！晶泰分子胶管线公布最新进展</title><link>https://www.xtalpi.com/%e6%b4%bb%e6%80%a7%e6%8f%90%e5%8d%87%e8%b6%85%e5%8d%83%e5%80%8d%ef%bc%81%e6%99%b6%e6%b3%b0%e5%88%86%e5%ad%90%e8%83%b6%e7%ae%a1%e7%ba%bf%e5%85%ac%e5%b8%83%e6%9c%80%e6%96%b0%e8%bf%9b%e5%b1%95/</link><description>[XtalPi official Chinese RSS] 活性提升超千倍！晶泰分子胶管线公布最新进展
并结合计算和实验数据训练 AI 模型，对不同 E3-目标蛋白组合进行优先级判断。</description><pubDate>Tue, 06 Oct 2026 03:29:09 GMT</pubDate></item><item><guid isPermaLink="false">bioai:6337ca57-735c-4c8c-960f-a59fb5d20e62</guid><title>CellART: a unified framework for extracting single-cell information from high-resolution spatial transcriptomics.</title><link>https://pubmed.ncbi.nlm.nih.gov/42834230/</link><description>[PubMed / NCBI E-utilities] Recent advances in spatial transcriptomics (ST) have achieved subcellular spatial resolution, yet existing platforms either capture sparse transcript counts per spot or measure only a limited number of genes, complicating the extraction of comprehensive single-cell information. H</description><pubDate>Mon, 05 Oct 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:a4831ca7-c357-4ea9-9875-edafd3c605bf</guid><title>Iambic Submits IND for a Potentially Brain-Penetrant KIF18A Inhibitor IAM217, Its Second Wholly Owned AI-Discovered Candidate</title><link>https://www.iambic.ai/post/iambic-submits-ind-for-a-potentially-brain-penetrant-kif18a-inhibitor-iam217-its-second-wholly-owned-ai-discovered-candidate</link><description>[Iambic official newsroom] In nonclinical studies, IAM217 was observed to have potent and selective target engagement with a favorable drug-drug interaction profile IAM217 is believed to be brain penetrant, producing regression of established intracranial tumors in a preclinical model IAM217 was designed u</description><pubDate>Mon, 05 Oct 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:dcd7b9b0-c4ff-4455-b8fa-d638a6e97105</guid><title>Engineering human multi-organ tissue chip niches for drug absorption, distribution, metabolism, excretion and toxicity prediction.</title><link>https://pubmed.ncbi.nlm.nih.gov/42834083/</link><description>[PubMed / NCBI E-utilities] Predicting human drug responses in vitro remains a central challenge in drug development. Animal models have shown limited accuracy in recapitulating human drug absorption, distribution, metabolism, excretion and toxicity, which arise from coordinated activities across multiple o</description><pubDate>Mon, 05 Oct 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:769fc6b2-f4dd-4cef-a7c1-7a8a852bf06a</guid><title>晶泰生发管线进入 IND-enabling 阶段：以极低浓度，解锁温和高效生发新可能</title><link>https://www.xtalpi.com/%e6%99%b6%e6%b3%b0%e7%94%9f%e5%8f%91%e7%ae%a1%e7%ba%bf%e8%bf%9b%e5%85%a5-ind-enabling-%e9%98%b6%e6%ae%b5%ef%bc%9a%e4%bb%a5%e6%9e%81%e4%bd%8e%e6%b5%93%e5%ba%a6%ef%bc%8c%e8%a7%a3%e9%94%81%e6%b8%a9/</link><description>[XtalPi official Chinese RSS] 晶泰科技（ 2228.HK ）今日宣布，其面向脱发治疗的创新药管线正式进入 IND-enabling（临床试验申请前准备）研究阶段。围绕候选分子的药效、安全性、成药性及 CMC 等关键要素的系统研究已经展开，为临床试验申请奠定基础。 据国家卫健委此前发布的数据，我国脱发人群已超过 2.5 亿，平均每 6 人中就有 1 人受脱发困扰。更值得关注的是，脱发正在明显年轻化——越来越多的 90 后、甚至 00 后，在人生早期阶段就开始面临发际线后移、发量减少等问题。 国内毛发健康市场持续扩容， 2023 年规模 795.5 亿元，预计 2028 年增长至 116</description><pubDate>Fri, 02 Oct 2026 02:52:02 GMT</pubDate></item><item><guid isPermaLink="false">bioai:bb5b6e11-2069-42c3-ae6a-a802b4acc41e</guid><title>FLIGHTED: Inferring fitness landscapes from noisy high-throughput experimental data.</title><link>https://pubmed.ncbi.nlm.nih.gov/42826717/</link><description>[PubMed / NCBI E-utilities] Machine learning (ML) for protein design requires large protein fitness datasets generated by high-throughput experiments for training and benchmarking models. However, most models do not account for experimental noise inherent in these datasets, thereby harming model performance</description><pubDate>Fri, 02 Oct 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:5911449c-c83f-433c-a7b5-8261c1747581</guid><title>Roswell Park, Generate Biomedicines Launch Exclusive Clinical Trial of ‘Armored’ CAR T Therapy for Ovarian Cancer</title><link>https://generatebiomedicines.com/news-releases/roswell-park-generate-biomedicines-launch-exclusive-clinical-trial-of-armored-car-t-therapy-for-ovarian-cancer/</link><description>[Generate Biomedicines official news-release RSS] Engineered with AI to optimize T-cell function in solid tumors, GB-5267 is the first next-generation cell therapy for ovarian cancer to enter clinical trials First patient treated in new clinical trial underway at Roswell Park Treatment is designed to avoid pre-infusion chemother</description><pubDate>Thu, 01 Oct 2026 11:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:7c6392c6-ccca-4422-bb26-17a825356188</guid><title>CytoVI: deep generative modeling of antibody-based single cell data.</title><link>https://pubmed.ncbi.nlm.nih.gov/42816645/</link><description>[PubMed / NCBI E-utilities] Antibody-based single-cell technologies, such as flow cytometry, mass cytometry and CITE-seq, have become widely used in clinical diagnostics and basic research; however, their analysis is complicated by technical noise, batch effects, platform differences and restricted antibody</description><pubDate>Wed, 30 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:dec42363-a527-4a49-8755-d867a41daae6</guid><title>Function-preserving watermarking of AI-generated proteins.</title><link>https://pubmed.ncbi.nlm.nih.gov/42816606/</link><description>[PubMed / NCBI E-utilities] Generative artificial intelligence (AI) models are revolutionizing biology, with tools such as AlphaFold 3 and protein design models accelerating breakthroughs in protein structure prediction and the creation of new functional proteins 1 . Tracking and establishing the provenance</description><pubDate>Wed, 30 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:84b9f799-b837-400c-9247-260d8815fc6e</guid><title>寻找下一个“十亿美金分子”科学创业大赛：XtalPi Science，高校年轻人的第一个创业合伙人</title><link>https://www.xtalpi.com/%e5%af%bb%e6%89%be%e4%b8%8b%e4%b8%80%e4%b8%aa%e5%8d%81%e4%ba%bf%e7%be%8e%e9%87%91%e5%88%86%e5%ad%90%e7%a7%91%e5%ad%a6%e5%88%9b%e4%b8%9a%e5%a4%a7%e8%b5%9b%ef%bc%9axtalpi-science/</link><description>[XtalPi official Chinese RSS] 符合要求的参赛团队将获得 XtalPi Science 启动 Token 支持。 经过实验验证并进入临床转化的优秀分子，有机会获得超亿元级现金投资。 近日，哈佛大学计算生物学博士 Douglas Yao 因在自家车库中借助 AI 和实验机器人推进候选药物研发，引发了广泛关注。 过去需要团队、实验室和大量专业工具才能完成的部分科研工作，正在被 AI 和自动化工具重新组织。一个科学想法，也开始有机会被更快地设计、验证，并继续向前推进。 一个人也能做科研创业的时代，开始了。 如果一个人就能借助 AI，把一个科学想法一步步推向真实实验、落地成创新项目，那么高校里</description><pubDate>Tue, 29 Sep 2026 07:54:08 GMT</pubDate></item><item><guid isPermaLink="false">bioai:6434c106-393c-4c06-8c49-1c256aca0e68</guid><title>Building a new path to make medicines with AI</title><link>https://www.isomorphiclabs.com/articles/building-a-new-path-to-make-medicines-with-ai</link><description>[Isomorphic Labs official announcements] Listen: Spotify Apple Podcasts The Scale Solving all disease is one of the most monumentally complex challenges in science. Today, for the first time, AI provides the foundation to make it achievable. Tackling the frontier of human health is our mission, and it means identifying </description><pubDate>Tue, 29 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:8b0ebefb-d8c8-4947-995c-8043af9acb75</guid><title>Estonia and Owkin Partner to Advance Sovereign AI in Biology</title><link>https://www.owkin.com/newsfeed/estonia-and-owkin-partner-to-advance-sovereign-ai-in-biology</link><description>[Owkin official newsroom] Tallinn / Paris, 29 September 2026 – Estonia, led by the National Applied Research Center Metrosert, and Owkin today announced a collaboration to advance sovereign AI in biological and health research, working with Metrosert, Estonia’s state-owned applied research organisation. T</description><pubDate>Tue, 29 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:1338d248-fc40-4383-9eb2-c2428e0760aa</guid><title>Absci Announces Preliminary Details for Phase 2 STORYLINE™ Trial of AI-Designed Antibody ABS-201™ for Endometriosis</title><link>https://investors.absci.com/news-releases/news-release-details/absci-announces-preliminary-details-phase-2-storylinetm-trial-ai</link><description>[Absci official news RSS] Phase 2 trial to assess efficacy and safety in participants with surgically-diagnosed endometriosis and moderate to severe endometriosis-associated pain (EAP) Anticipated initiation of Phase 2 trial in mid-2027 VANCOUVER, Wash. and NEW YORK, Sept. 24, 2026 (GLOBE NEWSWIRE) -- Abs</description><pubDate>Thu, 24 Sep 2026 12:00:49 GMT</pubDate></item><item><guid isPermaLink="false">bioai:5a1a71e3-3b98-412d-a779-2d89aae0cb64</guid><title>晶泰向美国FDA提交 IND申请，AI + 机器人加速创新药物迈向临床</title><link>https://www.xtalpi.com/%e6%99%b6%e6%b3%b0%e5%90%91%e7%be%8e%e5%9b%bdfda%e6%8f%90%e4%ba%a4-ind%e7%94%b3%e8%af%b7%ef%bc%8cai-%e6%9c%ba%e5%99%a8%e4%ba%ba%e5%8a%a0%e9%80%9f%e5%88%9b%e6%96%b0%e8%8d%af%e7%89%a9%e8%bf%88/</link><description>[XtalPi official Chinese RSS] 晶泰科技（2228.HK）今日宣布，其自主研发的肠道限制性高选择性pan-TRK抑制剂KQTD-126已完成临床前研究， 并已向美国食品药品监督管理局（FDA）提交新药临床试验（IND）申请 。 这款抑制剂主要用于治疗肠易激综合征（IBS）及炎症性肠病（IBD）相关慢性肠道疼痛。有望为长期受慢性肠道疼痛困扰的患者提供兼顾疗效与安全性的新选择，也为“高活性、高选择性与组织靶向性难以兼得”这一药物设计经典难题提供了新的解题思路。 这一成果充分体现了晶泰科技“AI + 机器人自主实验”平台在药物研发中的独特价值：AI在活性、选择性与组织分布等多维参数之间协同寻</description><pubDate>Thu, 24 Sep 2026 09:02:06 GMT</pubDate></item><item><guid isPermaLink="false">bioai:1e8c7e66-5329-49f7-9c61-b30dd745faf6</guid><title>晶泰AI4S助力艾滋病治愈获关键进展：候选药物高效提名PCC</title><link>https://www.xtalpi.com/%e6%99%b6%e6%b3%b0ai4s%e5%8a%a9%e5%8a%9b%e8%89%be%e6%bb%8b%e7%97%85%e6%b2%bb%e6%84%88%e8%8e%b7%e5%85%b3%e9%94%ae%e8%bf%9b%e5%b1%95%ef%bc%9a%e5%80%99%e9%80%89%e8%8d%af%e7%89%a9%e9%ab%98%e6%95%88/</link><description>[XtalPi official Chinese RSS] 近日，晶泰科技 （2228.HK）孵化企业艾玮泰在 HIV （艾滋病） 药物研发上取得新进展， 其围绕 HIV 关键靶点开发的长效治疗项目已确定临床前候选药物（PCC）。 这一进展把针对 HIV 关键靶点的结构认知落地为真实的药物研发成果，有望帮助患者摆脱长期服药的现状，也为探索艾滋病彻底治愈带来新的可能。 这一成果充分展现了结构研究与 AI 药物设计协同的价值：依托靶点结构信息开展定向筛选，提升实验的针对性，并结合实验反馈迭代优化分子。 对于其他同样拥有结构基础与实验验证条件的靶点，这套研发路径同样具备借鉴价值，有助于降低早期研发的试错成本，加速候选药</description><pubDate>Tue, 22 Sep 2026 08:39:29 GMT</pubDate></item><item><guid isPermaLink="false">bioai:ab050bdc-a633-4361-9668-1cf777901fd1</guid><title>AbbVie and Iambic Announce Collaboration to Accelerate AI-driven Drug Discovery</title><link>https://www.iambic.ai/post/abbvie-and-iambic-announce-collaboration-to-accelerate-ai-driven-drug-discovery</link><description>[Iambic official newsroom] − Multi-year partnership applies Iambic’s AI platform to speed up small molecule drug discovery in immunology, neuroscience and oncology − Companies to pursue programs with both first-in-class and best-in-class potential, combining Iambic’s molecular superintelligence platform wi</description><pubDate>Mon, 21 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:e2723a49-0d5b-4b73-9ee2-346ea40a8ffe</guid><title>Iambic Launches Enchant v3 – Molecular Superintelligence Designed to Advance End-to-End Drug Discovery &amp; Development</title><link>https://www.iambic.ai/post/iambic-launches-enchant-v3---molecular-superintelligence-designed-to-advance-end-to-end-drug-discovery-development</link><description>[Iambic official newsroom] Enchant v3 is the next generation of Iambic’s multimodal transformer model, trained using 41 billion parameters on more than 6,000 molecular properties across 16 biomedical data modalities Enchant v3 ingests diverse datasets, deploys a mixture-of-experts architecture, and tests m</description><pubDate>Mon, 21 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:452e5e0e-d5f2-4f82-b0d4-20c8787e4fe6</guid><title>The Virtual Biotech: A multi-agent AI framework for therapeutic discovery and development.</title><link>https://pubmed.ncbi.nlm.nih.gov/42752167/</link><description>[PubMed / NCBI E-utilities] Drug development requires evidence integration across biological scales and modalities, but relevant tools are fragmented. We introduce the Virtual Biotech, an organization of artificial intelligence (AI) agents modeled on a drug-development company, with agentic divisions spanni</description><pubDate>Thu, 17 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:e7645dfc-c408-44b0-8493-6dbddea03a83</guid><title>Owkin to License K Pro and Multimodal Data to Servier to Advance Oncology Research</title><link>https://www.owkin.com/newsfeed/owkin-to-license-k-pro-and-multimodal-data-to-servier-to-advance-oncology-research</link><description>[Owkin official newsroom] NEW YORK &amp; PARIS - September 11, 2026 — Owkin, the agentic AI company pioneering the first automated AI scientist to revolutionize drug discovery and development, today announced an agreement with Servier, an independent international pharmaceutical group governed by a foundation</description><pubDate>Fri, 11 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:8244ffd9-0aaf-4bf9-beaa-c95d423dfb56</guid><title>AI做实验如何少走弯路？晶泰科技联合多所高校提出CARE框架</title><link>https://www.xtalpi.com/ai%e5%81%9a%e5%ae%9e%e9%aa%8c%e5%a6%82%e4%bd%95%e5%b0%91%e8%b5%b0%e5%bc%af%e8%b7%af%ef%bc%9f%e6%99%b6%e6%b3%b0%e7%a7%91%e6%8a%80%e8%81%94%e5%90%88%e5%a4%9a%e6%89%80%e9%ab%98%e6%a0%a1%e6%8f%90%e5%87%ba/</link><description>[XtalPi official Chinese RSS] 近期，晶泰科技联合麦吉尔大学、哈佛大学、UCLA 等高校研究团队，针对 AI 自主实验如何提升优质反应条件筛选命中率这一核心痛点，研究团队提出 CARE（Context‑Aware Ranking Evolution）多专家协同框架。 该框架面向高通量实验（HTE）的下一轮实验条件推荐，融合多专家模型、大语言模型与证据审计机制协同决策，通过多重评估约束，优先筛选出最具备验证价值的实验条件。 多项真实反应优化任务与公开基准测试显示， CARE 取得当前最优整体表现 。 从反复试错, 到更早找到优异条件 化学实验是药物发现中的关键环节。想找到更优的反应条件，</description><pubDate>Thu, 10 Sep 2026 10:14:16 GMT</pubDate></item><item><guid isPermaLink="false">bioai:d2bdfd75-f9dd-4cfa-b3e2-daef7179cdd1</guid><title>4D spatiotemporal landscape of mitochondrial phenotypes across cellular states unlocked through representation learning.</title><link>https://pubmed.ncbi.nlm.nih.gov/42721963/</link><description>[PubMed / NCBI E-utilities] Mitochondria are four-dimensional (4D: x, y, z, and time) organelles essential for cellular function. Characterizing their 4D phenotypic landscape across diverse cellular states requires both 4D imaging and analytical frameworks. We present MitoSpace, a self-supervised deep learn</description><pubDate>Thu, 10 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:61aa4aa3-b654-43c8-9037-18f2d7242fde</guid><title>全球首款口服LDHA抑制剂冲刺临床，AI+机器人构建新药研发范式</title><link>https://www.xtalpi.com/%e5%85%a8%e7%90%83%e9%a6%96%e6%ac%be%e5%8f%a3%e6%9c%8dldha%e6%8a%91%e5%88%b6%e5%89%82%e5%86%b2%e5%88%ba%e4%b8%b4%e5%ba%8a%ef%bc%8cai%e6%9c%ba%e5%99%a8%e4%ba%ba%e6%9e%84%e5%bb%ba%e6%96%b0%e8%8d%af/</link><description>[XtalPi official Chinese RSS] AI 与机器人参与研发的新药，正迈入临床验证新阶段。 日前，晶泰科技（2228.HK）孵化企业默达生物宣布：双方合作研发的全球首款口服 LDHA 抑制剂 MP-5342，已完成大鼠、犬 GLP 毒理研究及猴预毒理研究，同步完成炎症性肠病（IBD）全部临床前药效验证，正式进入 IND 申报冲刺阶段，计划近期启动临床Ⅰ期。 从靶点验证、化合物筛选，到多轮优化攻坚，再到接连闯过毒理与药效两大核心关卡，MP-5342 这款潜在首创新药的完整研发路径，背后正是晶泰科技“AI+机器人”研发平台的全链条支撑。 而它也是继 META-001-PH 之后，晶泰科技与默达生</description><pubDate>Wed, 09 Sep 2026 01:30:43 GMT</pubDate></item><item><guid isPermaLink="false">bioai:45ff0472-a9fe-4695-9495-476a839ea33b</guid><title>Sequence and structural determinants of efficacious de novo chimaeric antigen receptors.</title><link>https://pubmed.ncbi.nlm.nih.gov/42716964/</link><description>[PubMed / NCBI E-utilities] Advances in generative protein design using artificial intelligence (AI) have enabled the rapid development of binders against heterogeneous targets, including tumour-associated antigens. Despite extensive biochemical characterization, these novel protein binders have had limited</description><pubDate>Wed, 09 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:56991e1e-7099-4d8a-a57e-1f9cd5bd4ee6</guid><title>An operational perturbation proteomics-based virtual cell model.</title><link>https://pubmed.ncbi.nlm.nih.gov/42717098/</link><description>[PubMed / NCBI E-utilities] Artificial intelligence-empowered virtual cell models represent an emerging approach for in silico drug discovery 1-3 , yet most existing approaches lack large-scale, time-resolved perturbation proteomics data and interpretable frameworks for predicting therapeutic responses. Her</description><pubDate>Wed, 09 Sep 2026 00:00:00 GMT</pubDate></item><item><guid isPermaLink="false">bioai:8a3b7f6c-7c40-4884-8d03-52cdc93c549f</guid><title>Predicting brain morphogenesis via physics-transfer learning.</title><link>https://pubmed.ncbi.nlm.nih.gov/42717040/</link><description>[PubMed / NCBI E-utilities] Brain morphology emerges from the interplay of genetic programming and mechanical forces, yet its fractal-like folding patterns make quantitative analysis and prediction difficult, especially when labeled data are scarce. Here we introduce a theory-grounded physics-transfer learn</description><pubDate>Wed, 09 Sep 2026 00:00:00 GMT</pubDate></item></channel></rss>