David Kenneth Gifford

This study designs factorizable antibody libraries using neural networks, finding they outperform random and enumerated libraries in pMHC binding.

The paper introduces VelociRAPTR, a method to map TCR-antigen interactions by combining yeast-displayed TCR libraries with pMHC-displaying virus-like particles, revealing biased recognition patterns and enabling TCR engineering.

This study identifies 21 genes whose rare variants affect LDL-C levels by altering cholesterol uptake, using CRISPR screening and co-essentiality analysis.

The paper maps gene expression and chromatin accessibility changes in mouse spinal motor neurons during maturation, revealing dynamic regulatory elements and shared transcription factor motifs.

seqgra is a deep learning pipeline that generates data based on hypothesized regulatory models and identifies neural networks that recover these rules, analyzing performance under varying conditions.

The paper introduces spatzie, an R package and website for detecting significant transcription factor motif co-enrichment in enhancer-promoter interactions using ChIA-PET data.

The paper introduces two deep learning methods to identify cell-type-specific DNA sequence patterns that influence chromatin accessibility and gene expression.

The paper evaluates nine computational methods for identifying and ranking reprogramming factors in directed differentiation, finding that chromatin accessibility-based methods like DeepAccess and AME perform best.

This study introduces STARR-scan, a computational method that identifies transcription factor motifs in synthetic DNA constructs linked to enhancer activity, revealing new regulators of the APOBEC-3B gene.

The paper introduces MIAA, a method to measure DNA accessibility changes caused by synthetic oligonucleotide motifs, and identifies key transcription factor motifs influencing chromatin accessibility in stem cells and endoderm.

This paper introduces PRESCIENT, a generative model for inferring cell differentiation landscapes from single-cell time-series data, showing improved prediction of cell fate and perturbation responses.

IDR2D identifies reproducible genomic interactions by filtering spurious ones from chromatin interaction data.

The paper introduces CID, a computational method for chromatin interaction discovery that improves sensitivity and consistency compared to existing methods.

This paper introduces a novel KSM motif representation and KMAC method for de novo motif discovery, showing improved accuracy in predicting TF binding sites and regulatory variant effects.

CpGenie predicts allele-specific DNA methylation and identifies methylation quantitative trait loci using deep learning.