UCLA

A disease can be caused by multiple mechanisms including genetic mutations, proteomic abnormalities, or abnormal mRNA transcription. While the Human Genome Project aims to understand the genetic basis of complex diseases, not all relevant variants are druggable targets. Therefore, a network approach can be useful for understanding if druggable proteins are close to gene variants. Previous development of PathFX, a network algorithm that identifies relationships between drugs and diseases, has given us more comprehension on genes / proteins that might be responsible for drug effects on disease and side effect phenotypes. While PathFX was constructed using a combination of gene expression, gene mutation, proteomics information, and other resources, the goal of the project is to understand how much a drug-phenotype relationship is driven by genetics within the PathFX network. We quantify the genetic component by finding the overlap between the PathFX network and the GWAS catalog data. We discovered that drugs connected to three disease groups: psychological disorders, autoimmune diseases, and inflammatory bowel diseases have greater enrichment of genetic information than other drug-disease pairs. Our quantitative assessment may uncover considerations for new drug developments, given that a disease is more genetically or less genetically enriched.