AI Genomics Collective (AIGCT) database

AI Genomics CollecTive (AIGCT) is a platform for systematically evaluating ML/AI models of variant effects across the spectrum of genomics-based precision medicine. It consists of a database of variant effect data organized into categories based on the source of the data. We use the term, task, to refer to a category. The data is accessed via a python package, agct. This dataset consists of a series of csv files which comprise the agct database compiled into a tar file. The data is sourced from: dbNSFP version 5.0am, released January 1, 2025, dbNSFP, license: Deed - Attribution-NonCommercial-NoDerivatives 4.0 International - Creative Commons