Our new paper using machine learning to identify non-coding variant effects is on Bio-Rxiv!

We’re excited to share this collaboration with the Tewhey Lab, along with the Sabeti lab and massive efforts by John Butts and Stephen Rong. In it, we introduce MPAC, a machine learning framework trained on MPRA data to predict the regulatory effects of non-coding variants genome-wide. MPAC enables scalable, accurate identification of causal, pathogenic, and cancer-associated regulatory variants, offering a valuable resource for functional genomics and variant interpretation. Read the preprint here!