Pathway and Network Integration of Cancer Genomics and Clinical Data

点击翻译按钮获取中文摘要

PROJECT SUMMARY Cancer genomics projects have successfully cataloged many of the frequent genomic, epigenetic, and gene expression alterations that drive cancer progression. However, these initial projects have also demonstrated that both the types and targets of genomic aberrations are incredibly heterogeneous, reflecting the large diversity of perturbations in the cellular machinery that promote tumor growth and metastasis. As cancer sequencing efforts expand to determine the molecular basis of additional phenotypes such as drug resistance or exceptional responders, novel methods to integrate data from multiple genomic characterization platforms across combinations of alterations in pathways and interaction networks are essential. We propose to build a Genome Data Analysis Center (GDAC) focused on pathway analysis. Our GDAC will integrate data from multiple genome characterization platforms, and use several computational approaches to identify combinations of genomic aberrations and downstream expression changes that distinguish clinical phenotypes. We will employ algorithms that utilize information about known pathways and/or biological interaction networks, as well as other approaches that analyze statistical patterns of mutual exclusivity and co-occurrence between alterations and clinical variables. We will combine the discovered pathways with knowledge of drugs and their targets to identify novel interventions in individual patients. Finally, we will augment the computational analyses with a web platform for interactive visualization and annotation of discovered pathways. This human-in-the-loop system will accelerate the annotation of mutations, pathways, and interventions and provide a dynamic ecosystem linking cancer genomics datasets to new and existing literature. By combining rigorous computational and statistical approaches with human-in-the-loop annotation, the proposed GDAC will facilitate the translation of multi-platform genome characterization data to clinical application.