新闻动态  News

查看更多  >>

  • 2024-12-25
    2024年12月21日,CMIS 2024第三届中国医药数智峰会在上海万豪虹桥大酒店圆满落幕!北京望石智慧科技有限公司被授予“2024年度医药行业AI智能新药创研先锋奖”,以肯定其以多模态 AI 分子生成大模型为撬点,帮助药物研发企业围绕IP整合数据、认知及工具,建立更快更好的药物研发新范式的努力与成果。望石创始人&CEO周杰龙同时受邀参与本次会议,并就“小分子创新药早期研发的数智化”进行分享,介绍了AI时代下医药行业的变化、望石对于当前小分子药物创新困局的思考、以及通过AI大模型辅助药物研发的可能性。   ✎ 主题演讲分享  CMIS 2024第三届中国医药数智峰会中,望石创始人&CEO周杰龙就“小分子创新药早期研发的数智化”这一话题进行了分享。从小分子创新药现状出发,揭示了行业内卷及创新困局的表象下,隐藏的数据、认知及工具无法很好串联从而无法被直接使用并转化为生产力的问题。接着周总详细介绍了望石针对这些问题提出的解决方案: 针对当前小分子创新困难、同质化内卷严重的...
  • 2024-11-15
    2024 年 9 月,望石智慧再次获得北京市科委、中关村管委会“ AI+ 健康协同创新培育项目”大额科技经费支持,课题为“融合电⼦密度的多模态 AI ⼩分⼦药物设计平台研发和应⽤”。此课题拟进一步挖掘电子密度中分子动态信息、溶剂信息等业界缺失的关系到“分子与目标蛋白如何相互影响”的关键信息,同时通过融合多模态迭代现有大模型。望石智慧希望通过该药物设计平台推动药物研发路径从“依赖于分⼦库的筛选已知小分子”升级为“空口袋生成小分子并通过人机交互不断优化筛选分子”的设计路径。  2024 年 10 月 9 日,AlphaFold 被授予诺贝尔化学奖,以表彰其通过创新性的引入已有的氨基酸残基共进化数据以及其对应的 MSA 表征来表⽰蛋⽩分⼦,进⽽推动困扰⼈类科学界 50 年的蛋⽩三维结构预测。与 AlphaFold 解决问题的思路类似,望石智慧在世界范围内首次创造性的把晶体学电子密度的拓扑性质作为蛋白分子和药物分子间/内相互作用的表示方法引入了 AI 模型,近乎完美的获得了一个可以表示物理性质...
  • 2024-04-03
    由广州实验室、美国化学会、美国化学会广东分会、广州国际生物岛集团有限公司主办,广州实验室病原体结构与临床应用创新研究院、广州国际生物岛集团有限公司、深圳理工大学合成生物学院承办的ACS创新药物研究和转化研讨会于2024年4月11日-12日在广州实验室召开,望石智慧副总裁黄博博士作为特邀报告人参会,并做了题为《大分子实验密度在AI辅助药物分子设计方面的应用》的学术报告。 报告综述了望石团队以大分子实验电子密度为数据切入口、以分子生成模型核心、以药物分子设计和筛选为应用场景的技术研发成果。6年来,望石团队深耕药物研发行业所特有的、尚未被充分利用的生物大分子实验电子密度数据,挖掘数据中蕴含的分子间相互作用信息、溶剂分布信息、以及分子的平均构象中所包含的构象动态信息,并且把这些用于分子生成模型的训练过程,相关学术成果发表于Nature,Nature Machine Intelligence,Nature Communications,Communications Chemistry, JCIM (封面),ACS Omega (封面) 等。 在报告中,黄博博士还着重强调了望石团队分子...
  • 2024-01-19
    分子生成是AI助力小分子新药研发的核心技术,理想能力的生成模型可以带来早期研发流程的重塑,并撬动巨大的商业价值。截至目前,业内依然没有达到工业标准使用的模型。 望石从成立起即始终专注于分子生成技术的开发。2024年1月15日,望石的研究团队在期刊《Nature Machine Intelligence》发表了题为《Generation of 3D molecules in pockets via a language model》的研究论文,并随文上线了学术版服务 (https://sw3dmg.stonewise.cn)。这是望石智慧第三代分子生成模型,也是3D分子生成模型的v2.0版本 (Lingo3DMol)。模型在分子生成关键指标——信息不泄漏情况下的已报道活性分子的复现、分子-口袋结合打分,以及分子构象方面均有优异表现。  2020年底,针对ligand-based场景,望石智慧发布了首代以骨架跃迁和衍生为主的2D生成模型(成果 | JCIM: AIScaffold基于深度学习的在线骨架衍生工具),该模型在生成分子的新颖性上有出色表现,帮助多家国内外药企在BIC项目上完成了专利的突破和项目进度的赶超。2022年,望石研究团...
  • 2023-12-12
    2023年12月11日,望石智慧研究团队作为共同第一作者与中国科学院物理所姜道华研究员团队和中国科学院生物物理研究所赵岩研究员团队在Nature发表了题为“Transport and inhibition mechanisms of human VMAT2”的研究论文,揭示了VMAT2在运输单胺底物过程中的构象变化及转运机制。 VMAT2是大脑中最重要的囊泡单胺转运蛋白,负责将5羟色胺、多巴胺、肾上腺素、去甲肾上腺素和组胺等神经递质转运到囊泡中储存,以便受到外界刺激后释放单胺神经递质。目前在临床上,VMAT2作为治疗高血压、亨廷顿舞蹈症等运动障碍、精神性焦虑的药物靶点。利血平(Reserpine, RES)和丁苯那嗪(Tetrabenazine, TBZ)是两种经典的VMAT2抑制剂。从20世纪50年代起,利血平被广泛用于治疗高血压。丁苯那嗪在临床上用于治疗亨廷顿舞蹈症等多动性运动障碍,年销售额达到10亿美元。尽管许多研究揭示了VMAT2的生物学和药理学性质,但是对于VMAT2的底物转运机制及药物分子的抑制机制仍不清楚。 VMAT2分子量仅为56 kDa, 利用冷冻电镜解...
  • 2023-12-08
    最近,望石团队在《Life Science Allian》上发表了题为《Patient-specific analysis of co-expression to measure biological network rewiring in individuals》的学术成果。 差异共表达(Differential Co-expression,以下简称DCE)网络分析方法被认为是利用算法处理基因数据,进而分析潜在疾病机制或制定个性化治疗方案的有用方法之一。常见的DCE网络分析方法通常基于多个样本的平均差异网络,忽视了个体患者之间的特异性数据。这种方法的局限性阻碍了对个体的深入理解和应用。 望石团队提出了一种新的模型Cosinet——一种DCE网络框架下的,单样本网络重连度量化工具(a DCE-based single-sample network rewiring degree quantification tool)。该方法能够利用基因表达数据确定单个样本的基因共表达模式和参考条件的相似程度,结合网络分析和统计方法,量化单个基因DCE的差异,从而解决了DCE网络分析在个体水平上的应用问题。 图1:Cosinet方法的流程 通过两个乳腺癌数据集进行验证,Cosinet能够识别个体患者之间基因共表达...
共 44 条,共 8 页

学术进展  Academic Progress

查看更多  >>

  • 2023-02-28
    Li, Y., Yuan, T., Huang, B. et al.  Nat Commun 14, 1030 (2023). https://doi.org/10.1038/s41467-023-36766-9 Abstract The sodium channel NaV1.6 is widely expressed in neurons of the central and peripheral nervous systems, which plays a critical role in regulating neuronal excitability. Dysfunction of NaV1.6 has been linked to epileptic encephalopathy, intellectual disability and movement disorders. Here we present cryo-EM structures of human NaV1.6/β1/β2 alone and complexed with a guanidinium neurotoxin 4,9-anhydro-tetrodotoxin (4,9-ah-TTX), revealing molecular mechanism of NaV1.6 inhibition by the blocker. The apo-form structure reveals two potential Na+ binding sites within the selectivity filter, suggesting a possible mechanism for Na+ selectivity and conductance. In the 4,9-ah-TTX bound structure, 4,9-ah-TTX bin...
  • 2022-10-09
    Cell. October 04, 2022 DOI:https://doi.org/10.1016/j.cell.2022.09.037 Liming Yan, Yucen Huang,  Ji Ge, Zhenyu Liu, Pengchi Lu, Bo Huang, Shan Gao, Junbo Wang, Liping Tan, Sihan Ye, Fengxi Yu, Weiqi Lan, Shiya Xu, Feng Zhou, Lei Shi, Luke W. Guddat, Yan Gao, Zihe Rao, Zhiyong Lou Summary Decoration of cap on viral RNA plays essential roles in SARS-CoV-2 proliferation. Here we report a mechanism for SARS-CoV-2 RNA capping and document structural details at atomic resolution. The NiRAN domain in polymerase catalyzes the covalent link of RNA 5’ end to the first residue of nsp9 (termed as RNAylation), thus being an intermediate to form cap core (GpppA) with GTP catalyzed again by NiRAN. We also reveal that triphosphorylated nucleotide analogue inhibitors can be bonded to nsp9 and fi...
  • 2022-09-15
    Yibo Li, Jianxing Hu, Yanxing Wang, Jielong Zhou, Liangren Zhang, and Zhenming Liu J Chem Inf Model 2020 Jan 27;60(1):77-91. doi: 10.1021/acs.jcim.9b00727. Epub 2019 Dec 20. Abstract The ultimate goal of drug design is to find novel compounds with desirable pharmacological properties. Designing molecules retaining particular scaffolds as their core structures is an efficient way to obtain potential drug candidates. We propose a scaffold-based molecular generative model for drug discovery, which performs molecule generation based on a wide spectrum of scaffold definitions, including Bemis-Murcko scaffolds, cyclic skeletons, and scaffolds with specifications on side-chain properties. The model can generalize the learned chemical rules of adding atoms and bonds to a given scaffold. The generated compounds were evaluated by molecular docking in DRD2 targets, and...
  • 2022-09-15
    Liming Zhao, Mengchen Pu, Huting Wang, Xiangyu Ma, and Yingsheng J. Zhang J Chem Inf Model. 2022 Sep 7. doi: 10.1021/acs.jcim.2c00616. Abstract In recent years, machine learning (ML) models have been found to quickly predict various molecular properties with accuracy comparable to high-level quantum chemistry methods. One such example is the calculation of electrostatic potential (ESP). Different ESP prediction ML models were proposed to generate surface molecular charge distribution. Electrostatic complementarity (EC) can apply ESP data to quantify the complementarity between a ligand and its binding pocket, leading to the potential to increase the efficiency of drug design. However, there is not much research discussing EC score functions and their applicability domain. We propose a new EC score function modified from the one originally develop...
  • 2022-09-15
    Sci Rep. 2022 Sep 6;12(1):15100. doi: 10.1038/s41598-022-19363-6. Lvwei Wang, Rong Bai, Xiaoxuan Shi, Wei Zhang, Yinuo Cui, Xiaoman Wang, Cheng Wang, Haoyu Chang, Yingsheng Zhang, Jielong Zhou, Wei Peng, Wenbiao Zhou & Bo Huang Abstract We report for the first time the use of experimental electron density (ED) as training data for the generation of drug-like three-dimensional molecules based on the structure of a target protein pocket. Similar to a structural biologist building molecules based on their ED, our model functions with two main components: a generative adversarial network (GAN) to generate the ligand ED in the input pocket and an ED interpretation module for molecule generation. The model was tested on three targets: a kinase (hematopoietic progenitor kinase 1), protease (SARS-CoV-2 main protease), and nuclear receptor (vi...
  • 2022-06-06
    Jiangtao Zhang, Yiqiang Shi, Junping Fan, Huiwen Chen, Zhanyi Xia, Bo Huang, Juquan Jiang, Jianke Gong, Zhuo Huang, Daohua Jiang Abstract: Voltage-gated sodium (NaV) channels initiate action potentials. Fast inactivation of NaV channels, mediated by an Ile-Phe-Met motif, is crucial for preventing hyperexcitability and regulating firing frequency. Here we present cryo-electron microscopy structure of NaVEh from the coccolithophore Emiliania huxleyi, which reveals an unexpected molecular gating mechanism for NaV channel fast inactivation independent of the Ile-Phe-Met motif. An N-terminal helix of NaVEh plugs into the open activation gate and blocks it. The binding pose of the helix is stabilized by multiple electrostatic interactions. Deletion of the helix or mutations blocking the electrostatic interactions completely abolish...
共 22 条,共 4 页
  • toolbar
    返回顶部