Large-Scale Machine Learning for Genome and Transcriptome Analysis
This team targets the development of large-scale machine learning and statistical algorithms to combine different layers of multimodal genomic data computationally to form complex models, which incorporate expert biological knowledge and predict biological outcomes based on the models in a context-dependent manner. The goal is to help clinicians and biologists gain new insights and to better understand disease mechanisms and to improve diagnosis and treatment.
Proteomics is the simultaneous, detailed study of all of the proteins made by an organism. The research in this team develops new ways to examine very large sets of results from proteomics experiments and to use data about protein and gene variations derived from previous experimental results to inform the analysis of new results. Issues of clinical concern are central themes, such as reproducibility, quality control and confidence levels.
Pharmacogenomics and Precision Medicine
Genomic factors play an important role in disease susceptibility and therapeutic treatment outcomes. By harnessing large-scale genomic datasets, and applying computational analyses to these data, this research aims to improve our understanding of how genetic variation contributes to human disease and the treatment thereof. These data will guide the development of individualized treatments for rare and common disease and the identification of novel drug targets. This research makes use of genome- and transcriptome-wide analyses, polygenic risk scores and drug repurposing analyses.