Anshul Bharat Kundaje

The paper identifies off-target effects in CRISPR-Cas9 screens for non-coding elements and shows that strict filtering with specificity scores improves the identification of essential regulatory regions.

The paper introduces a neural network architecture for optimizing regulatory DNA sequences to enhance transcription factor binding and gene expression.

The paper introduces MPRA-DragoNN, a CNN model to predict regulatory DNA activity from MPRAs, identifies TF motifs, and explores variant effects in lipid-associated loci.

The paper analyzes cell cycle dynamics in human pluripotent stem cells primed for differentiation using RNA-sequencing and identifies PI3K signaling's role in early differentiation states.

The paper predicts genome-wide chromatin accessibility across cell types by integrating DNA sequence and gene expression data using neural networks.

gkmexplain provides a fast and accurate method to interpret gkm-SVM models of regulatory DNA sequences, outperforming existing methods in identifying predictive patterns and mutation impact scores.

This study analyzes epigenomic and transcriptomic changes across mouse tissues with age, identifying age-related immune pathway activation and retroviral sequence upregulation.

This paper evaluates transcriptome-wide association studies (TWAS) for causal gene prioritization, showing its strengths and limitations in identifying causal genes at GWAS loci.

The paper evaluates reproducibility and quality measures for Hi-C data using real and simulated datasets, showing that specialized methods outperform simple correlation analysis.

The paper identifies a conserved DNase-I footprinting pattern across human mtDNA, suggesting a regulated protein-DNA organization involved in mtDNA transcription.

Kipoi provides a framework for standardized, reusable machine learning models in genomics, enabling tasks like variant effect prediction and model benchmarking.

The paper introduces Umap and Bismap tools to assess genome and methylome mappability, revealing significant non-uniquely mappable regions in both unmodified and bisulfite-converted genomes.

The paper introduces DFIM, a method to identify epistatic feature interactions in neural network models of regulatory DNA sequences, validated on simulated and in vivo data.

Deep learning rivals or beats baselines across imaging, health records, genomics and drug discovery but is not yet transformative; scarce labels, privacy and opacity remain hurdles.

GenomeDISCO measures concordance between chromosome conformation capture contact maps by smoothing them with random walks, distinguishing biological replicates from different cell types.

The paper examines regulatory variation in iPSCs and differentiated cells, finding lower individual variation in iPSCs and identifying shared chromatin regions for some cell type-specific QTLs.

This study uses ATAC-seq in C. elegans to identify dynamic chromatin accessibility changes, revealing novel enhancers and a regulatory factor involved in development.

This paper uses HiChIP to map enhancer-promoter interactions in primary human cells, linking disease-associated DNA variants to their target genes.

Omni-ATAC improves ATAC-seq by reducing background noise and enabling analysis of frozen tissues, revealing disease-associated DNA element activities in human brain structures.

PyBoost and HiBoost are parallelized Python implementations of GeneClass for learning gene regulatory networks from paired transcription factor and motif data, applied to regeneration and differentiation studies.

This paper introduces a separable fully connected layer for deep learning in genomics, showing improved performance and biological consistency over conventional models.

The paper introduces a convolutional denoising algorithm, Coda, to improve histone ChIP-seq data quality by learning to map suboptimal to high-quality data.

Standard genomic CNNs can score a sequence and its reverse complement differently; four weight-sharing layer types remove that inconsistency and are reported to train faster.

The paper describes using Segway software to annotate genomes with chromatin data from ChIP-seq experiments, providing a semi-automated workflow for interpretation.