Anshul Bharat Kundaje

The paper introduces SMFNet and PRISM to analyze how transcription factor assemblies influence chromatin accessibility and identifies sequence-encoded principles of combinatorial TF assembly.

The paper develops deep learning models to predict transcription initiation patterns from DNA sequences across human tissues and cancers.

The paper introduces Watershed-DL, a framework integrating sequence-to-omics predictions and multi-omics data to prioritize rare variants, finding it increases prioritized variants by 61.4% and identifies candidate disease alleles.

This paper introduces Swap-seq, a method to systematically delete and quantify the effects of enhancers, silencers, and topological elements on gene expression in human cells.

The paper maps and redesigns the MYBPC3 promoter to enhance expression and treat haploinsufficient hypertrophic cardiomyopathy.

This study compares divergent and unidirectional transcriptional regulatory elements, finding that unidirectional elements are younger and less constrained, with some transcription factors modulating their directionality.

The paper shows that conserved motif vocabularies constrain genome access in cell type evolution across species.

A fully orthogonal IL-2 receptor platform isolates 36 cytokine receptors in T cells; receptor-specific STAT combinations specify distinct fates, and STAT3-dominant programs give curative antitumor responses.

The paper introduces scE2G models to predict enhancer-gene regulatory interactions from single-cell data and demonstrates their performance across multiple cell types and perturbations.

The paper develops an integrative framework to classify and interpret promoter-centric chromatin interactions by combining Hi-C, ChIA-PET, and functional genomic data, identifying a distinct class of architectural promoter-enhancer interactions with context-dependent regulatory roles.

The paper introduces APB-Display, a method for high-throughput in vitro protein library production and analysis, quantifying binding affinities and expression levels of thousands of variants.

The paper characterizes regulatory variants in human neurons using lentiMPRA and shows that variant effects are influenced by regulatory element activity and sequence context, not population frequency.

A 350 bp masked DNA language model trained on ENCODE cCREs with a frequency-domain regularizer recovers TF motifs zero-shot, scores regulatory variants, improves ChromBPNet inputs, and guides sequence design.

This paper introduces PISA, a tool for interpreting sequence-to-function neural networks, enabling the discovery of nucleosome-positioning motifs and understanding cis-regulatory code through MNase-seq data.

BenchRep-T evaluates nine T-cell repertoire methods across four tasks, finding simple baselines competitive with complex models and no single method dominates all tasks.

The paper introduces a framework for multimodal chemical screens combining RNA, protein, and epigenomic readouts, revealing compound-specific mechanisms and off-target effects in T cells.

Bio-BLIP is a multimodal architecture that integrates DNA, genes, proteins, and text to improve variant interpretation without task-specific fine-tuning, achieving better performance in variant prioritization and target gene prediction.

TwinC predicts trans-chromosomal DNA contacts using DNA sequence and provides insights into their functional mechanisms.

The paper characterizes atherosclerosis trajectories in vascular smooth muscle cells using single-cell data and identifies causal transcriptomic and epigenomic mechanisms of CAD risk.

This study shows that many premalignant colorectal polyps contain multiple genetic lineages, indicating polyclonal origins rather than single-cell expansion.

This study creates a transcriptomic and spatial atlas of human arterial segments, revealing cell-type and segment-specific heterogeneity linked to vascular disease risk.

The paper uses a single-cell multi-omic atlas and deep learning to identify cis-regulatory elements and genetic drivers of heart development and disease.

The paper predicts missing chromatin conformation contact maps across cell types and assays, exploring how 3D genome architecture varies.

This paper develops beCasKAS, a method to directly detect off-target base editor activity in primary cells, revealing distinct off-target profiles and enabling risk stratification of edits.

The paper presents a high-resolution atlas of transcriptional regulatory elements in human tissues and disease states using PRO-cap and deep learning models to prioritize noncoding variants.

The paper identifies causal genetic mechanisms at CAD GWAS loci using CRISPR screens and chromatin analysis, revealing enhancer-gene links and ancestry/sex-specific risk factors.

This study shows that cancer cell phagocytosis by macrophages induces an anti-inflammatory gene regulatory program, promoting immunosuppression and tumor growth.

The paper introduces a high-resolution single-molecule chromatin accessibility mapping method to quantify transcription factor occupancy and co-accessibility in human cell lines.

The paper maps the mutagenic sensitivity of human enhancers in vivo by systematically mutating enhancer sequences and using predictive modeling to identify critical nucleotides.

The paper introduces Variant-FlowFISH to measure the effects of DNA sequence variants on gene expression, identifies regulatory mechanisms, and benchmarks predictive models.