Anshul Kundaje

TL;DR: It is shown that the average accessibility of a genomic region across training contexts can be a surprisingly powerful predictor and novel strategies for training models are employed to enhance genome-wide prediction of shared and context-specific chromatin accessible sites across cell types.

TL;DR: The data and analyses presented here suggest that study of GWAS transcription factors may be a highly useful approach to identifying disease gene interactions and thus pathways that may be relevant to complex disease etiology.

TL;DR: GkmExplain is proposed: a computationally efficient feature attribution method for interpreting predictive sequence patterns from gkm-SVM models that has theoretical connections to the method of Integrated Gradients and consistently outperform deltaSVM and ISM at identifying regulatory genetic variants from gmmVMs models of chromatin accessibility in lymphoblastoid cell-lines.

TL;DR: This study uses ATAC-seq to identify chromatin accessibility changes in a whole animal, C. elegans, and identifies EOR-1 as a unique transcription factor that may regulate chromatin dynamics during development, and demonstrates the power of using whole organism Chromatin accessibility to identify novel regulatory regions in complex systems.

TL;DR: Predictive properties in isolation are inspected, and meta‐analysis over the competing methods find that using chromatin accessibility and transcription factor binding as features in an ensemble of classifiers or regression models leads to the most accurate results.