Project 1: Machine Learning for Biological Network Analysis

Project description

The main objective of my project is to develop methods for network-based genome-wide association studies (GWAS) that combine computational efficiency, statistical power and interpretability, thereby enabling the discovery of biological pathways underlying complex phenotypic traits. GWAS aim to identify statistical associations among genetic variants, also referred to as single nucleotide polymorphisms (SNPs), and disease risk or other phenotypes. The identification of such associations can positively affect healthcare as it enables enhanced disease prevention, diagnosis and personalised treatment. To date, GWAS rarely make use of the rich knowledge regarding biological networks, such as protein-protein interaction or gene regulation networks, underlying the phenomenon of interest. Including such contextual and functional information, however, can help to increase the statistical power as well as interpretability in GWAS aimed at complex biological traits that do not follow Mendelian inheritance laws and are influenced by environmental factors.

Secondments

1. Establishing a conceptual connection between significance testing and causality discovery
   Host: Siemens Healthcare GmbH
   Planned date: February 2021 – April 2021
2. Applying network mining to biobank-scale datasets
   Host: University of Tartu
   Planned date: March 2022 – May 2022

Publications

Giulia Muzio*, Leslie O’Bray* and Karsten Borgwardt (* = equal contribution). (2020). Biological network analysis with deep learning. Briefings in Bioinformatics 2020, bbaa257. https://doi.org/10.1093/bib/bbaa257