Modelling complex biological systems with machine learning and integrative single-cell genomics

Abstract

The advent of genome sequencing revolutionized our understanding of human genetics. Recently, experimental advances have allowed for sequencing genetic material at the level of single cells, unmasking heterogeneity of living systems beyond the tissue level, and leading to a better mechanistic understanding of disease processes. Advances in sequencing methods capturing multiple data modalities at the single-cell level necessitate development of computational methods to analyze heterogenous and noisy data measured across individuals, timepoints, and disease states. This talk will offer a primer on single-cell sequencing, applications in cancer research, standard computational approaches to data analysis, and the development of machine learning and deep learning approaches to systems modelling using single-cell data. After an overview, the specific challenge of data integration in the field will be discussed, focusing on machine learning approaches and ongoing work in the Bo Wang Lab pertaining to this research area.