Using Interpretable Machine Learning and Network Systems Approaches to Uncover Molecular Mechanisms of Immune Regulation and Dysregulation

Presenter: Dr. Jishnu Das, Assistant Professor of Immunology and Computational Biology

Dr. Das's lab is currently working on using network systems and functional genomic approaches to perform multi-scale integration of genomic and epigenomic datasets with biological networks to identify molecular phenotypes underlying a range of immunological disorders. They also use statistical and interpretable machine-learning techniques to integrate multi-omic datasets (genomic, transcriptomic, proteomic, metabolomic and antibody-omic) and elucidate molecular mechanisms of immune regulation and dysregulation. Dr. Das's talk will discuss recent progress made on both fronts. 

Background Reading

Using machine learning to uncover immune correlates in infectious diseases

• Route of immunization defines multiple mechanisms of vaccine-mediated protection against SIV

• Mapping functional humoral correlates of protection against malaria challenge following RTS,S/AS01 vaccination

• Mining for humoral correlates of HIV control and latent reservoir size

Moving beyond predictive to interpretable machine learning approaches

• Latent Model-Based Clustering for Biological Discovery

Using network systems approaches to elucidate molecular phenotypes in genetic disease

• Extensive disruption of protein interactions by genetic variants across the allele frequency spectrum in human populations