Research areas

We work on algorithm development for homolog identification, multiple sequence alignment, phylogenetic tree construction, protein fold prediction, identification of domain boundaries (and novel domains), and detection of key amino acids, such as catalytic or binding pocket residues. Structure prediction methods have obvious implications on the inference of protein function, as protein function is mediated by protein fold. Because evolution conserves structure in order to conserve function, we integrate phylogenetic tree construction and subfamily identification into our protein structure and function prediction methods, to enable us to infer the changes produced in protein function and structure over the evolution of a protein superfamily.