Martin Ng’ang’a Muigano

LOL: a Python package for leakage-aware phenotype prediction and confounding diagnostics in GEO transcriptomics datasets

LOL identifies phenotype-associated patterns and detects confounding factors in GEO transcriptomics data through leakage-safe preprocessing and diagnostic checks.

Protein language modeling and supervised machine learning reveal the functional landscape of antimicrobial resistance genes in African metagenomes

This study uses protein language models and machine learning to analyze antimicrobial resistance genes in African metagenomes, revealing geographic variation and key sequence features distinguishing ARGs from non-ARGs.

Pangenome and comparative genomics analysis reveal patterns of quorum sensing and antimicrobial resistance in Salmonella Lineages from African Food Chains

This study analyzes Salmonella genomes from African food chains to identify patterns of antimicrobial resistance and quorum sensing, revealing lineage-specific genes and geographic diversity.