Reactome IDG portal: Pathway-based analysis and visualization of understudied human proteins

点击翻译按钮获取中文摘要

Project Summary Proteins function through interactions with other proteins and biological entities to form biological pathways inside and between cells. Targeted therapies are designed to mitigate or reverse malfunctions caused by mutations in proteins via providing drugs to recover normal biological pathways’ activities. Projecting understudied proteins in the context of biological pathways is a powerful way to infer potential functions of these proteins. Pathway-based approaches are now routinely applied in bioinformatics and computational biology data analysis and visualization. Pathway databases are essential for those approaches. During the past two decades, our team has been working together on building the Reactome knowledgebase, arguably the most popular and comprehensive open source biological pathway database, covering over half of human protein-coding genes and widely used in the research community. In this application, we propose to develop a Reactome IDG pathway portal, which will allow localization of understudied proteins in biological pathways, identifying likely interactions with better-known proteins in specific processes annotated in Reactome, pinpointing most effective drug targets via pathway modeling, thus generating testable predictions of molecular functions of these proteins in key domains of biology. Specially we will develop a web-based application to place understudied proteins in the context of Reactome pathways by importing a variety of data types collected in the IDG projects and other resources and then overlaying them onto the Reactome pathways by leveraging existing Reactome software tools (e.g. interaction overlay). Furthermore, we will develop a machine learning approach to predict functional interactions between understudied proteins and well-known Reactome annotated proteins and a Boolean network-based fuzzy logic modeling approach to integrate the scores produced from the machine learning approach to simulate the impacts of understudied proteins on pathways’ activities. We believe our approach will provide a unique and powerful approach to help the community to understand the contribution of the understudied proteins to cellular functions.