BMP1

摘要深度生成模型具有加速药物设计的巨大潜力。然而，由于数据有限，现有的生成模型常常面临泛化方面的问题，导致设计创新性较差，并且与看不见的目标蛋白之间往往存在不利的相互作用。为了解决这些问题，作者提出了一种相互作用感知的 3D 分子生成框架，该框架能够在目标结合口袋内进行相互作用引导的药物设计。通过分析生成的未见靶标配体的结合姿势稳定性、亲和力、几何图案、多样性和新颖性，对其性能进行了全面评估。通过利用蛋白质-配体相互作用的普遍模式作为先验知识，其模型可在训练集实验数据有限的情况下进行推广。此外，潜在突变选择性抑制剂的有效设计证明了作者的方法对基于结构的药物设计的适用性。方法作者的框架由两个主要阶段组成 (1) 相互作用感知条件设置和 (2) 相互作用感知 3D 分子生成 - 如图 1所示。图 1：作者提出的相互作用感知 3D 配体生成框架的概念化说明a第一阶段对蛋白质口袋进行分析，为每个蛋白质原子指定一个相互作用条件。b在第二阶段，DeepICL 根据预定的相互作用条件在蛋白质口袋内顺序添加配体原子。圆圈内的字母表示相互作用类型如下：氢键 ( H )、疏水相互作用 ( D )、盐桥 ( S ) 和π - π堆积 ( π )。第一阶段该框架的第一阶段旨在通过研究给定结合位点P的蛋白质原子来设置相互作用条件I。作者使用了四种类型的蛋白质-配体相互作用——氢键、盐桥、疏水相互作用和π - π堆积，并使用了源自 PDB 的PDBbind 2020 数据集进行模型训练。作者开发了一种蛋白质原子相互作用感知调节策略，将相互作用条件定义为一组蛋白质原子的附加相互作用类型one-hot向量，它指示原子是否可以参与特定的相互作用以及它在相互作用中的作用。蛋白质原子分为七类之一：阴离子、阳离子、氢键供体和受体、芳香族、疏水性和非相互作用原子。作者的策略是在本地建立交互条件，如图1b所示 ，在原子添加的每一步中仅考虑相邻的口袋原子。在这项工作中，作者主要通过两种策略确定口袋原子的相互作用类别，如图 1a所示。在生成阶段，有关受体如何与配体相互作用的信息常常不可用。因此，作者预先定义了相互作用类别的标准，以便作者可以通过分析它们来指定每个蛋白质原子的相互作用条件。由于作者不使用任何参考配体进行条件设置，因此将此条件称为无参考相互作用条件。例如，作者渲染 SMARTS 模式来确定氢键受体和供体。同时，在训练阶段存在蛋白质-配体复合物C的真实结构，因此作者从这些参考结构中提取相互作用条件。作者使用了蛋白质-配体相互作用分析仪 (PLIP)，通过分析蛋白质和配体的结合结构来识别蛋白质和配体之间的非共价相互作用。除了这两种策略之外，从实际的角度来看，人们可以根据对目标系统的了解，手动指定所需的交互条件。第二阶段在第二阶段中，根据口袋的 3D 背景和第一阶段的相互作用条件顺序生成配体中的原子。为此，作者设计了一种名为 DeepICL（深度交互感知条件配体生成模型）的深度生成模型，用于进行配体的逆向设计。如图1b所示 ，配体原子逐个添加，并且每一步中感兴趣的原子都会发生变化。该感兴趣的原子指示添加下一个原子的位置。因此，作者只考虑感兴趣原子C t的周围环境，以便重新定义局部相互作用条件It ，以馈送到当前的生成步骤t中。作者展示了作者在靶蛋白内部的两个分子生成任务（配体精加工和从头配体设计）的框架。前一项任务旨在精炼已知配体以提高其效力。后者旨在从头开始设计配体，提供适合结合口袋的多种分子。对于配体精细化任务，给出配体核心结构的结合姿势并将其用作初始状态。在从头配体设计任务中，人们可以手动选择口袋内的一个点作为起点。结果：相互作用感知条件配体设计的效果在药物设计过程中，构建特定的蛋白质-配体相互作用至关重要，因为它们可能与药物的效力和选择性直接相关。假设配体与之相互作用的结合位点是已知的，生成框架就能够设计可以有利地与这些位点相互作用的配体。为了对这些相互作用结合位点做出合理的猜测，作者利用了参考蛋白质-配体复合物的相互作用模式。图 2：相互作用感知条件配体精加工的示例a 标有给定相互作用条件的初始核心结构和结合袋表面。b具有最高相互作用相似性的原始配体和设计配体以及相应的相似值。c、d口袋与原始配体或设计配体之间相互作用的二维图。圆圈表示氨基酸残基，虚线表示相互作用。不同的颜色用于区分相互作用类型，其中多种颜色的圆圈对应于参与不止一种相互作用类型的残基。每个配体中突出显示用作初始结构的核心结构。e在参考和掩蔽条件下生成的配体相互作用相似性的分布。左：骨形态发生蛋白 1（BMP1，PDB ID：6bto），中：成纤维细胞生长因子 1（FGF1，PDB ID：3ud9），右：二氢叶酸还原酶（DHFR，PDB ID：1dis）。在数据部分描述的测试复合物中，作者选择了具有不同蛋白质-配体相互作用的复合物来证明相互作用感知调节的效果。作者选择了三种蛋白质，即骨形态发生蛋白1（BMP1）、成纤维细胞生长因子-1（FGF1）和二氢叶酸还原酶（DHFR）。从它们原始配合物的参考配体中，作者提取了初始核心结构，分别是氮杂双环[2.2.1]庚烷、2-(oxan-3-yloxy)oxane和苯。核心结构是根据作者的手动检查确定的，去除链和官能团以留下由单环或双环组成的最小结构，如图2a 所示。图 2说明了配体精心设计的三个示例，其中它们给定的相互作用条件在图2a中以斑块形式可视化 。在图 2b中，描绘了具有最高相互作用相似性的设计配体以及原始配体。在这里，相互作用相似度估算的是生成配体与原始配体之间相互作用模式的相似度。作者在方法部分介绍了交互相似性的精确定义。图2c、d分别描绘了具有最高相互作用相似性的原始配体和设计配体的相互作用图。对于第一种情况，即图2的左栏 ，作者的模型成功设计了氢键、π - π堆积和盐桥作为与不同基序（motifs）的给定相互作用条件。值得注意的是，图 2 d-1表明该模型可以生成噻吩环代替原来的苯环，与TYR68构建π - π堆积。这意味着该模型了解了形成π - π堆叠所需的芳香环的特征。尽管该模型在疏水性 PHE157 附近添加了脂肪族碳，但该距离略大于疏水相互作用的阈值。对于另外两种情况，DeepICL 还成功设计了配体，这些配体表现出与原始配体高度相似的相互作用模式，同时生成与原始配体不同的基序（motifs）。作者进一步证明了作者的相互作用感知调节策略在再现相互作用模式方面的效果，这些模式是作为阐述每个配体核心结构的条件而给出的。作为一种消融设置，作者屏蔽了交互条件，以排除推理过程中有关参考交互模式的信息。作者为模型提供了一个与原始交互条件大小相同的零向量作为屏蔽交互条件。作者比较了图2e中参考或屏蔽条件生成的两个配体组之间相互作用相似性的分布 。它清楚地显示了每种情况下两种分布之间的显着差异，其中使用相应参考条件精心设计的配体实现了更高的相互作用相似性。因此，作者证明其框架是高度可控的，通过对现有配体的阐述为配体提供所需的相互作用模式。结论作者提出了一种相互作用感知的 3D 分子生成框架。与仅依赖结合结构信息的模型不同，蛋白质与配体相互作用模型是通用的，与蛋白质配体对无关，因此具有更高的普适性。作者重点研究了四种类型的蛋白质配体相互作用--疏水相互作用、氢键、盐桥和π-π堆叠--的通用模式，并通过综合分析未见过的靶标的设计配体，证明了其方法的普适性。说明结合蛋白质-配体相互作用先验知识可以更好地推广基于机构的药物设计。本研究表明，采用适当的先验知识,可以提高深度生成模型在数据有限的各科学领域中的普适性。参考资料：Zhung, W., Kim, H. & Kim, W.Y. 3D molecular generative framework for interaction-guided drug design. Nat Commun 15, 2688 (2024). https://doi.org/10.1038/s41467-024-47011-2版权信息本文系AIDD Pro接受的外部投稿，文中所述观点仅代表作者本人观点，不代表AIDD Pro平台，如您发现发布内容有任何版权侵扰或者其他信息错误解读，请及时联系AIDD Pro (请添加微信号sixiali_fox59)进行删改处理。本文为原创内容，未经授权禁止转载，授权后转载亦需注明出处。有问题可发邮件至sixiali@stonewise.cn关注我，更多资讯早知道↓↓↓

Peptide-level insights from deep, unbiased proteomics enable protein variant analysis of patient samples, revealing potential non-small cell lung cancer biomarkersREDWOOD CITY, Calif., March 29, 2023 (GLOBE NEWSWIRE) -- Seer Inc. (NASDAQ: SEER), a life sciences company commercializing a disruptive new platform for proteomics, today announced a new scientific publication in PLOS One demonstrating the unmatched utility of the Proteograph Product Suite to uncover novel insights in human health and disease with deep, unbiased proteomics. Scientists from Seer in collaboration with Luis Diaz, M.D. from Memorial Sloan Kettering Cancer Center found previously unknown associations between four protein isoforms and non-small cell lung cancer (NSCLC) progression, constituting potential novel, disease-relevant biomarkers, or therapeutic targets. Importantly, the identified NSCLC-associated protein isoforms comprise structural variants of proteins that arise from distinct but related mRNAs produced from the same gene through the process of alternative splicing. Different protein isoforms can have distinct roles in biology, influencing disease predisposition and progression. Notably, discovery and quantification of such protein variants in a complex biological sample requires deep, unbiased interrogation of the proteome at peptide-level resolution. “These findings demonstrate that distinct protein isoforms may differentially contribute to diverse biological mechanisms and to the pathogenesis of cancers, potentially paving the road to identify new diagnostic markers or new therapeutic targets,” said Dr. Diaz, Head of the Division of Solid Tumor Oncology at Memorial Sloan Kettering Cancer Center. “What is especially encouraging here is that these protein isoforms were detected in plasma, a readily accessible sample type, enabling cancer detection and monitoring through liquid biopsy evaluation of patients.” The researchers found that the short protein isoform of bone morphogenetic protein 1 (BMP1) occurred more frequently in NSCLC patients compared with healthy participants, and with even greater abundance in patients with late-stage cancer, suggesting that this isoform of BMP1 may play a role in NSCLC progression. BMP1 is involved in collagen processing, and the short isoform of the protein lacks the ability to release collagen, which could impact the tumor microenvironment. The study analyzed 188 plasma proteomes from NSCLC patients and controls to identify disease-associated protein isoforms. In addition to BMP1, the analysis revealed three other protein isoforms with significant differential behavior in NSCLC when compared to healthy controls: complement component 4a (C4a), complement component 1r (C1r) and lactate dehydrogenase B (LDHB). “This study, which found disease-associated protein isoforms in NSCLC, could only have been achieved using an unbiased peptide-level approach that provides a deeper, more nuanced view of the human proteome,” said Asim Siddiqui, Senior Vice President of Research at Seer. “These findings demonstrate the power of unbiased proteomics to discover novel biology, especially in the context of cancer, that may otherwise be missed by genomics or targeted proteomics.” Further research is necessary to validate the utility of these protein isoforms as novel biomarkers for NSCLC, or even new therapeutic targets. Seer’s Proteograph Product Suite enables proteomics studies with an unprecedented combination of speed, scale, depth, and breadth of data, allowing an unbiased interrogation of the proteome to allow studies not previously possible. Seer’s proprietary engineered nanoparticles deliver reproducible performance across samples, labs, and experiments, providing peptide level information that is key to identifying protein variants. The accompanying Proteograph Analysis Suite offers cloud-scalable software for proteomic data analysis, visualization, and generation of biological insights. The Proteograph Product Suite makes it easy to add unbiased, deep, rapid proteomics studies at scale to any lab. About SeerSeer is a life sciences company developing transformative products that open a new gateway to the proteome. Seer’s Proteograph™ Product Suite is an integrated solution that includes proprietary engineered nanoparticles, consumables, automation instrumentation and software to perform deep, unbiased proteomic analysis at scale in a matter of hours. Seer designed the Proteograph workflow to be efficient and easy to use, leveraging widely adopted laboratory instrumentation to provide a decentralized solution that can be incorporated by nearly any lab. Seer’s Proteograph Product Suite is for research use only and is not intended for diagnostic procedures. For more information, please visit www.seer.bio. Forward-Looking StatementsThis press release contains “forward-looking statements” within the meaning of the Private Securities Litigation Reform Act of 1995, as amended. Such forward-looking statements are based on Seer’s beliefs and assumptions and on information currently available to it on the date of this press release. Forward-looking statements may involve known and unknown risks, uncertainties and other factors that may cause Seer’s actual results, performance, or achievements to be materially different from those expressed or implied by the forward-looking statements. These statements include but are not limited to statements regarding access to proteins, discovery and quantification of protein variants and biological insights, identification and utility of protein isoforms as novel biomarkers and therapeutic targets, deep unbiased interrogation of the proteome at the peptide level, the role of BMP1 in NSCLC pathogenesis and treatment resistance, the role of BMP1 in collagen processing and the tumor microenvironment, the ability of the Proteograph Product Suite to enable proteomics studies with an unprecedented combination of speed, scale, depth, and breadth of data. These and other risks are described more fully in Seer’s filings with the Securities and Exchange Commission (“SEC”) and other documents that Seer subsequently files with the SEC from time to time. Except to the extent required by law, Seer undertakes no obligation to update such statements to reflect events that occur or circumstances that exist after the date on which they were made. Media Inquiries:Elizabeth Eatonpr@seer.bio Investor Inquiries:Carrie Mendivilinvestor@seer.bio

Researchers have identified several genetic variants associated with increased risk of cancer in children with non-chromosomal birth defects, such as congenital heart disease and defects of the central nervous system. While the risk of developing cancer is not as high as children with chromosomal birth defects, it is significantly higher than children with no birth defects at all, and the findings may provide a basis for early detection in these understudied patients. Researchers from Children's Hospital of Philadelphia (CHOP) have identified several genetic variants associated with increased risk of cancer in children with non-chromosomal birth defects, such as congenital heart disease and defects of the central nervous system. While the risk of developing cancer is not as high as children with chromosomal birth defects, it is significantly higher than children with no birth defects at all, and the findings may provide a basis for early detection in these understudied patients. The findings were recently published in the journal Biomarker Research. Children with birth defects are more likely to develop cancer, and that increased risk of cancer persists into adulthood. Prior studies have demonstrated that children with chromosomal birth defects, such as Down syndrome and Klinefelter syndrome, are more than 11 times more likely to be diagnosed with cancer than children without any birth defects. However, children with non-chromosomal birth defects are up to 2.5 times more likely to be diagnosed with cancer than those without birth defects. With birth defects of any kind occurring in 1 in every 33 births in the United States each year, that increased risk implicates a significant number of children. The underlying genetics of non-chromosomal birth defects have not been studied in great detail. Researchers at the Center for Applied Genomics (CAG) at CHOP wanted to determine what molecular mechanisms were at play and potentially identify genetic clues that could lead to early identification of cancer in these patients. "We assembled one of the largest pediatric oncology and birth defects projects in children as part of the Gabriella Miller Kids First program project, which helps to uncover new insights into childhood cancer and structural birth defects," said Hakon Hakonarson, MD, PhD, director of the CAG at CHOP and senior author of the study. "With this partnership, we sought to identify functional molecular pathways based on mutations we identified as part of this study." In this study, researchers used data obtained from whole genome sequencing of blood samples from 1,653 individuals without chromosomal abnormalities that were acquired from the Kids First Data Resource Center. These samples included 541 birth defect probands -- the first person in a family to receive genetic counseling or testing for hereditary risk of a disease -- with at least one type of malignant tumor, 767 birth defect probands without malignant tumors, and 345 healthy family members who are parents or siblings of the aforementioned probands. Additionally, once variants were identified, whole genome sequencing data from 40 birth defect probands from outside the data resource center, including 25 patients with at least one type of cancer, were used to further validate the study. The study identified thousands of variants of interest, including 119 genes with at least two variants in coding regions -- regions of the gene that will eventually be transcribed and translated into proteins to carry out essential functions -- and 478 genes with at least 20 variants in their non-coding regions. Five genes in particular -- AXIN2, BMP1, CR1, ERBB2, and RYR1 - are associated with birth defects and increased risk of cancer. Additionally, the researchers built a deep learning model to assess the variants of interest identified in the Kids First cohort when compared with the 40 validation samples and found that they had achieved approximately 75% accuracy, with even greater accuracy for variants that were associated with non-coding regions. Further detailed analysis of this data could identify genes and noncoding mutations that not only result in specific birth defects but also identify which types of cancer they are more prone to and whether the risk is more pronounced in childhood or adulthood. "While more research is needed to delve into the variants of interest we identified, this study represents a critical step toward the earlier detection of cancer children with non-chromosomal birth defects," Hakonarson said. The sequencing data was provided through the Gabriella Miller Kids First Pediatric Research Program consortium (Kids First), supported by the Common Fund of the Office of the Director of the National Institutes of Health. The study was supported by Institutional Development Funds from CHOP to the CAG.