Submitted by Nathan Salomonis (Cincinnati Children's Hospital, USA) on Jan 17 2017
Platform: ngs – Illumina HiSeq 2500 (Homo sapiens)
Pubmed: 29662164
Summary Explaining the genetics of many diseases is challenging because most associations localize to regulatory regions. We present a novel computational method for discovering disease-driving mechanisms acting across multiple disease-associated, non-coding genomic regions. Application to a matrix of 213 phenotypes and 1,544 transcription factor (TF) binding datasets identifies 2,264 significant associations for hundreds of TFs in 92 phenotypes, including prostate and breast cancers. Strikingly, nearly half of the systemic lupus erythematosus risk loci are occupied by the Epstein-Barr virus EBNA2 protein and 24 human TFs, revealing an important gene-environment interaction. EBNA2-anchored associations also exist in multiple sclerosis, rheumatoid arthritis, inflammatory bowel disease, type 1 diabetes, juvenile idiopathic arthritis, and celiac disease. Instances of allele-dependent DNA binding and downstream effects on gene expression at plausibly causal autoimmune variants support a genetic mechanism of pathogenesis centered on EBNA2. Our results nominate mechanisms operating across disease risk loci, suggesting new paradigms of disease origins.
ID | Title | Cell Type | Timepoint | Reported Virus | Virus Species | Exclusion Reason |
---|---|---|---|---|---|---|
GSM2460844 | Ramos EBV neg | Burkitt lymphoma cell line ![]() ![]() |
none | Uninfected | ||
GSM2460845 | Ramos EBV pos | Burkitt lymphoma cell line ![]() ![]() |
unknown | EBV | Human gammaherpesvirus 4 |