David Burstein

FAMUS improves protein annotation by using contrastive learning to compare query sequences against multiple profiles, outperforming existing tools in KEGG and PANTHER annotations.

This paper develops a transformer-based classifier to detect artificially edited bacterial genomes by analyzing gene family contexts, achieving accurate identification without relying on specific marker genes.

The paper identifies YfjB as a host reprogramming factor that enables plasmid dissemination by altering host transcription and metabolism.

This study explores using reduced amino acid alphabets with sub-word tokenization in protein language models, showing improved efficiency with minimal performance loss.

The paper identifies context-dependent factors affecting CRISPR-Cas9 editing efficiency across species using endogenous editing data and shows that repair outcomes are conserved.

B-PPI is a bacterial PPI prediction model that outperforms human-centric models on bacterial data with higher AUPRC and F1 scores.

This study shows phage-encoded homing endonucleases reduce bacterial immunity by inhibiting GAPS4, a DNase defense system, thereby enhancing phage fitness.

This paper presents a CRISPR-Cas12a assay for rapid, on-site detection of Bactrocera zonata and Ceratitis capitata, demonstrating high specificity and sensitivity for pest monitoring.

PCQC selects optimal principal components for clustering single-cell RNA-seq data with imbalanced classes by focusing on variance explained tails, outperforming existing methods.