Petety V. Balaji

This paper predicts fold-switching in proteins using handcrafted and language model features, finding binary classification outperforms 3-class and handcrafted features avoid sequence memorization.

The paper evaluates BLASTp and profile HMMs for fine-grained functional annotation, finding BLASTp struggles with remote homologs while HMMs improve sensitivity but may miss some family members due to functional divergence.

This study evaluates nine clustering algorithms for unsupervised protein function annotation using pseudo-amino acid composition features, finding some methods effective for functional family segregation but less so for fine-grained differences.

The paper uses sequence composition features to classify paralogous proteins and identifies key amino acids driving classification.

The paper identifies sequence motifs near the pyranose binding site that confer reaction specificity in fold type I aminotransferases, enabling classification of their substrate specificity.

The paper discusses methods to customize functional homolog searches using profile HMMs and evaluates bit score thresholds through case studies.

GlycoPathDB compiles experimentally characterized monosaccharide biosynthesis pathways and enzymes, providing annotations and sequence homologs for evolutionary and functional studies.

The paper hypothesizes that the glycan alphabet is not universal, showing variation in monosaccharide biosynthesis pathways across bacterial and archaeal genomes.

The paper investigates the prevalence of rhamnose biosynthesis pathways across genomes and metagenomes, finding clade-specific distribution and strain-level variation in prokaryotes.