Evgeny Z. Kvon

The paper designs tissue-specific enhancers for mouse embryos using deep learning models trained on chromatin accessibility data and validated enhancers.

The paper identifies a conserved enhancer cluster regulating efnb2 expression in the dorsal retina and links it to human retinal thickness variation.

The paper identifies a conserved cis-acting element, REX, that enables extreme long-range enhancer-promoter interactions in mammals by increasing the genomic interaction range of enhancers.

The paper shows that poised enhancers are susceptible to non-coding variants causing gene activation and disease, using the ZRS and autism-associated variants as examples.

This paper introduces dual-enSERT, a tool for rapidly measuring enhancer variant activity in live mice, and shows it can identify shared genetic mechanisms in non-coding variant pathogenesis.

Learning and removing Tn5/DNase-I sequence bias lets a small base-resolution CNN (ChromBPNet) recover TF motifs and footprints from accessibility data and rival Enformer at predicting chromatin QTL effects.

The paper examines 3D enhancer-promoter interactions during mammalian development, finding that most enhancers bypass neighboring genes and show tissue-specific conformations with increased interactions upon activation.