Deep-learning Integration of Histopathology and Proteogenomics at a Pan-cancer Level - Resubmission

泛癌水平上组织病理学和蛋白质基因组学的深度学习整合 - 重新提交

• 批准号: 10606760
• 负责人: Joshua Wang
• 金额: $ 4万
• 依托单位: NEW YORK UNIVERSITY SCHOOL OF MEDICINE
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-03-01 至 2025-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10606760
• 关键词: Address; Antibodies; Antigen Presentation; Architecture; Automobile Driving; Biological; Biomedical Research; Biopsy; Cancer Prognosis; Cellular Morphology; Clinic; Clinical; Collaborations; Communities; Complement; Complete Blood Count; Computing Methodologies; DNA Sequence Alteration; Data; Data Set; Decision Making; Diagnosis; Disease; Disease-Free Survival; Disparate; Dreams; Endometrial Carcinoma; Genes; Genetic Transcription; Glioblastoma; Hematology; Histologic; Histology; Histopathology; Image; Immunohistochemistry; Immunophenotyping; Individual; Intuition; KRAS2 gene; Knowledge; Knowledge Discovery; Learning; Malignant Neoplasms; Malignant neoplasm of lung; Measurement; Medical; Methods; Mitotic; Modeling; Molecular; Molecular Profiling; Morphology; Multiomic Data; Mutation; National Cancer Institute; Neural Network Simulation; Nuclear; Outcome; PIK3CA gene; Pathologic; Pathologist; Pathology; Pathway interactions; Patients; Pattern; Performance; Process; Prognosis; Proteins; Proteomics; Regulatory Pathway; Research; Resolution; Role; Stains; TP53 gene; Tissues; Training; Trust; Tumor Biology; Tumor Pathology; Tumor stage; Urine; Validation; Visual; Visualization; Work; ZFHX3 gene; cancer therapy; cancer type; clinical practice; clinical predictors; convolutional neural network; data integration; deep learning; deep learning model; density; diagnostic biomarker; diagnostic tool; diagnostic value; differential expression; illness length; imaging platform; immune cell infiltrate; improved; interest; machine learning model; machine learning prediction; multiple omics; novel; personalized diagnostics; predict clinical outcome; predictive signature; proteogenomics; reconstitution; tool; transcriptomics; tumor; tumor heterogeneity; tumorigenesis

PROJECT SUMMARY Discovering and understanding novel pathological features that correlate with proteogenomics is crucial in improving and streamlining cancer prognosis and treatment. Currently, biomedical research efforts to predict cancer outcomes rely on sequencing approaches not readily accessible in a clinical setting. Instead, clinicians frequently rely on histopathology images to visually assess for aberrant changes to tissue morphology. Subsequently, a tool to infer clinical and molecular signatures directly from histopathology images would harness the power of omics research with the feasibility of image-based diagnosis. Histopathology imaging data is still vastly underutilized in the quest to better understand tumor biology, largely because of inadequate tools for analysis and data integration. To identify pan-cancer hallmarks and conserved pathways of tumorigenesis, I therefore propose to develop multi-resolution deep convolutional neural network (CNN) models across 10 different cancer types and predict clinical annotations, histology outcomes, and critical mutations based on tumor histopathology images (Aim 1). Through our lab’s collaboration with the National Cancer Institute’s Clinical Proteomic Tumor Analysis Consortium (CPTAC), we have access to multi-omics, clinical, and histopathologic data obtained from 1,602 patients. In addition, to understand the biological mechanisms driving morphology changes, I propose to develop computational methods that integrate transcriptomic and proteomic expression datasets with imaging to facilitate pathway-level knowledge discovery (Aim 2). Our proposal is the first to correlate expression perturbations with morphology patterns and identify enriched canonical pathways directly from histopathology images. Importantly, this proposal aims to connect scientific efforts of biomedical research with the diagnostic tools of clinicians to expand diagnostic power and improve clinical practice.

项目摘要 发现和理解与蛋白质基因组学相关的新的病理学特征是至关重要的， 改善和简化癌症预后和治疗。目前，生物医学研究努力预测 癌症结果依赖于在临床环境中不容易获得的测序方法。相反，临床医生 通常依赖于组织病理学图像来视觉评估组织形态的异常变化。 随后，直接从组织病理学图像推断临床和分子特征的工具将利用 组学研究的力量与基于图像的诊断的可行性。 在寻求更好地了解肿瘤生物学方面，组织学成像数据仍然大大未得到充分利用， 因为分析和数据整合工具不足。为了识别泛癌症的特征， 因此，我建议开发多分辨率深度卷积神经网络 (CNN)10种不同癌症类型的模型，并预测临床注释、组织学结局和关键 基于肿瘤组织病理学图像的突变（Aim 1）。通过我们实验室与国家 癌症研究所的临床蛋白质组肿瘤分析联盟（CPTAC），我们可以获得多组学， 从1,602名患者中获得的临床和组织病理学数据。此外，要了解生物 驱动形态变化的机制，我建议开发计算方法， 转录组学和蛋白质组学表达数据集与成像，以促进路径水平的知识发现 (Aim 2）。我们的建议是第一个将表达扰动与形态学模式相关联， 直接从组织病理学图像中丰富了经典途径。重要的是，该提案旨在连接 生物医学研究的科学努力与临床医生的诊断工具，以扩大诊断能力， 改善临床实践。