Gene modules associated with human diseases revealed by network analysis

Within biological networks, genes associated with human diseases likely map to modules whose identification facilitates etiology studies but remains challenging. We describe a systematic approach to identify such disease-associated gene modules. A gene co-expression network was constructed using GTEx dataset and assembled into 652 gene modules. Screening these modules identified those with disease genes enrichment for obesity, cardiomyopathy, hypertension, and autism, which revealed the pathways involved in their pathogenesis. Using mammalian phenotypes derived from mouse models, potential disease candidate genes were identified from these modules. Also analyzed were epilepsy, schizophrenia, bipolar disorder, and depressive disorder, revealing shared and distinct disease modules among brain disorders. Thus disease genes converge on modules within our network, which provides a general framework to dissect genetic basis of human diseases.