Development of machine learning software to quantitatively map telomere induced senescence in tissue sections during aging

开发机器学习软件以定量绘制衰老过程中组织切片中端粒诱导的衰老图谱

• 批准号: 10376395
• 负责人: Joao Passos
• 金额: $ 55.65万
• 依托单位: MAYO CLINIC ROCHESTER
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-21 至 2023-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10376395
• 关键词: 3-Dimensional; Address; Age; Aging; Atlases; Biological; Biopsy; Cell Aging; Cell Nucleus; Cells; Chronic; Collaborations; Communities; Computer software; Computing Methodologies; Cytometry; DNA Damage; Data; Detection; Development; Disease; Emerging Technologies; Fluorescence Microscopy; Functional disorder; Generations; Hour; Human; Image; Impairment; Intervention; Knowledge; Laboratories; Length; Machine Learning; Maps; Measurement; Measures; Mediating; Methods; Mitochondria; Morphologic artifacts; Mus; Nature; Phenotype; Play; Process; Proteins; Research; Resources; Role; Signal Transduction; Signaling Protein; Spatial Distribution; Stains; Suspensions; Technology; Time; Tissues; Treatment Efficacy; age related; base; clinical practice; deep learning; extracellular; human tissue; imaging modality; innovation; innovative technologies; insight; learning network; microscopic imaging; multiplexed imaging; neural network; next generation; prognostic tool; response; senescence; single cell technology; telomere; tissue resource

PROJECT SUMMARY/ABSTRACT Cellular senescence is a pillar of aging, acting as a key driver of aging and age-related diseases. Telomeres play a major role in cellular senescence. When telomeres become critically short or damaged, they elicit a DNA damage response (DDR) that drives senescence. Our laboratory developed sophisticated methods to detect senescent cells in tissues based on the co-localization between telomeres and the DDR. Furthermore, we recently developed SenoQuant, a software that simplifies the measure of Telomere associated foci (TAF), reducing quantification time from weeks to hours. We now propose to apply emerging technologies such as machine learning and deep learning to map TAF more accurately and robustly in human tissue sections. This will address several challenges inherent to TAF analysis, including nuclei detection, staining artifacts, and quantification time. Furthermore, in collaboration with the Tissue Mapping Centers we will tailor SenoQuant to the analysis of specific human tissues. Finally, we propose to integrate TAF with multiplexed imaging methods such as Imaging Mass Cytometry (IMC), allowing the detection of multiple senescence markers in tissues simultaneously. We anticipate that this technology will greatly advance the spatially-resolved mapping of senescent cells in human tissues and will be a great resource for the aging and cell senescence community.

项目概要/摘要 细胞衰老是衰老的一个支柱，是衰老和与年龄相关的疾病的关键驱动因素。 端粒在细胞衰老中发挥着重要作用。当端粒变得非常短或 受损后，它们会引发DNA损伤反应（DDR），从而导致衰老。我们的实验室 开发了基于共定位的复杂方法来检测组织中的衰老细胞 端粒和 DDR 之间。此外，我们最近开发了 SenoQuant，一个软件 简化了端粒相关灶 (TAF) 的测量，减少了定量时间 几周到几小时。我们现在建议应用机器学习等新兴技术 深度学习可以在人体组织切片中更准确、更稳健地绘制 TAF。这将 解决 TAF 分析固有的几个挑战，包括细胞核检测、染色伪影、 和量化时间。此外，与组织绘图中心合作，我们将 定制 SenoQuant 以分析特定的人体组织。最后，我们建议整合TAF 使用多重成像方法，例如成像质量细胞术 (IMC)，可以进行检测 同时检测组织中的多个衰老标志物。我们预计这项技术将 极大地推进了人体组织中衰老细胞的空间分辨绘图，并将成为 老龄化和细胞衰老界的重要资源。