Developing an integrative bioinformatic approach for the selection of therapeutic targets

点击翻译按钮获取中文摘要

The exploding volume of genomic and structural data related to human biology, and the continuous development of computational tools to interrogate this data generate novel opportunities to investigate human biology. The goal of this research program is to apply this paradigm to the target discovery process. We will develop novel methodologies to identify protein targets that both have clear genetic linkage to disease, and are structurally druggable. Our long-term goal is to expand methodologies and lessons learnt from this exercise to further integrate at the atomic and genetic level the molecular circuitry that underlies biological processes. Embracing an open science concept will facilitate the dissemination of our findings across and beyond our existing network of collaborators, and maximize our impact.******Selection of a target protein is the first step in most drug discovery programs, and determines whether the program will succeed or fail to deliver a new drug to patients. If the three-dimensional structure of the target is not chemically tractable, the program is doomed before it started. Equally important is the necessity to establish a causative and essential role of the targeted protein in the disease pathway. To increase the success rate in drug discovery, the target selection process needs to be better informed.******Big Data that started flooding the biomedical community is a transformative source of information. The global goal of my research program is to integrate structural and genomics data to investigate human biology, with a focus on the regulation of transcription in health and disease. In this proposal, while focusing on the regulation of gene expression, an area of biology that is central to multiple disease pathways but has resisted most drug discovery efforts, we will interface large databases of chromosomal aberrations from cancer patients and synthetic lethality screens (where deletion of Gene A causes dependency on Gene B for survival) to identify tumor-specific genetic vulnerabilities, and reveal proteins that should be targeted by drugs. We will also analyze the druggability and structural diversity of binding pockets at the surface of thousands of high-resolution structures of human proteins to identify targets that physically can be targeted by drugs. We will map proteins that can, that should, and that are targeted by drugs in the context of protein networks and signalling pathways, and identify novel high-priority targets that so far have been neglected. Students working on this project will acquire a rarely combined set of expertise in structural bioinformatics and computational genomics, and will benefit from a cross-disciplinary, open space environment composed of structural biologists, biochemists, cell biologists. They will also be able to share with and learn from the vibrant local bioinformatics community of downtown Toronto.