The Governing Council of the University of Toronto

Published purpose

Precise discovery on the causal associations between phenotype (such as human complex diseases) and underlining genetic variations is one of the fundamental questions in life sciences, and has an essential role in translational medicine. To date, many studies have been conducted to infer disease-relevant genes/biomarkers, but only a limited number of signals have been experimentally confirmed to be causative. The cell heterogeneity of the complex tissues is a major known factor contributing to this limitation. In this project, we aim to precisely identify genetic-phenotype associations in a cell-subtype specific manner by combining massively parallel single-cell-based CRISPR knockdown or knockout screening and advanced AI technologies, such as deep generative models. In particular, activities will focus on applying and developing AI methods to mine the real information behind the massive single cell data and infer the precise genetic-phenotype signals. Research will also make use of AI technologies to design gene and cell therapeutic interventions to target the identified signals.