A software platform facilitating community analyses of genetic datasets for complex disease
Year: 2016
Type: other
Abstract: While genome wide association studies (GWAS) have identified numerous variants of modest effect on complex disease risk, converting these findings into biological insights has proved challenging. The large quantities of data produced by different GWAS and sequencing studies too often remain disparate, and are inaccessible to the broader biomedical community. Here we present a new platform to facilitate analyses of genetic data for complex disease. Our platform, which powers the Type 2 Diabetes Knowledge Portal (T2DKP), includes software for aggregating and harmonizing genetic association study results, along with highly customizable genotype-phenotype association analyses across data sets, all while protecting sensitive patient data. The platform allows users to perform: Forward genetic analyses: visualize the top association statistics from any collection of GWAS or sequencing studies in either tabular or graphical form, and customize searches across datasets based on arbitrary association criteria. As an example, we demonstrate the identification of candidate association signals proposed to affect T2D through effects on cholesterol efflux from the liver based on searches across glycemic and lipid datasets. Reverse genetic analyses: return a list of association statistics for any variant, gene, or region across all datasets, with harmonized directions of effect, optionally drilling down to explore ethnic- or cohort-specific effects. As an example, we demonstrate the identification of secondary protective effects on T2D for a variant implicated in coronary artery disease (CAD) risk based on a simple search for the variant. Custom association analysis: specify a phenotype, and optionally a property for stratifying samples, along with sample filters, and phenotype or property covariates, then dynamically perform association tests on one or more variants. As an example, we demonstrate the prediction of a potential method of action for a T2D association based on interactive analysis of patients with a specific disease subtype. The T2DKP is built upon this platform, and is freely and publicly available at www.type2diabetesgenetics.org. The T2DKP draws over 100 users a day, each of whom interacts with the portal for over six minutes on average. Portals for other complex genetic diseases, such as stroke and heart attack, have been demonstrated using the platform. This open source framework may be useful to the larger community.
Institution: Broad Institute
Language: English
Subfield: Genetics
Domain: Life Sciences
Open Access status: gold
OpenAlex ID: W2565319564