Mahmoud ElHefnawi

PathoBERT improves bacterial pathogenicity prediction from short reads by combining a DNABERT encoder with attention-based feature extraction, achieving better performance than existing methods.

The paper performs an in-silico analysis of HCC omics data to show microRNAs can act as both tumor suppressors and oncomirs through complex regulatory interactions.

This paper introduces a systematic auditing framework to identify biases in machine learning applications in biology, showing that biases can hinder model generalizability and lead to poor performance when training data lacks sufficient signal.

The paper reports whole-genome sequences of nine Egyptian females with clinical data, identifying SNPs and mtDNA mutations associated with obesity and CPEO, and analyzing genetic admixture patterns.

The paper presents a large-scale cancer pharmacogenomic screen with 11,576 experiments on 910 drug combinations across 85 cancer cell lines, enabling crowd-sourced prediction of synergistic drug pairs and biomarkers.

The paper uses feature selection techniques to identify essential proteins for learning and memory in a mouse model of Down syndrome, finding higher accuracy with deep feature selection than traditional methods.