About

Orthology inference using Hidden Markov Models (OrthoHMM) was developed as part of Jacob L. Steenwyk’s post-doctoral work.

Inferring orthology (shared ancestry through speciation rather than duplication) is a major challenge in bioinformatics. This challenge is exacerbated when sequence divergence is high – for example, among organisms that began diverging dozens of millions of years ago – which makes remote homology difficult to detect.

A popular approach to identify orthology utilizes graph theory. Specifically, an all-to-all comparison of sequence similarity is conducted using BLAST or DIAMOND and the resulting similarity scores are used to cluster groups of highly similar sequences together. The resulting groups of sequences are putative orthologs; thus, these genes are termed orthologous groups of genes (or orthogroups).

OrthoHMM uses profile Hidden Markov Models for protein similarity search and family refinement. Its phylogenetic mode additionally reconciles inferred gene trees. Benchmark results depend on the dataset, output semantics and configuration; the publication analyses do not establish general superiority over other tools. In particular, gains in paralog separation can accompany losses in homolog coverage. The repository’s publication results retain these trade-offs, failures and uncertainty estimates.

The Developers

OrthoHMM is developed and maintained by Jacob L. Steenwyk and Thomas J. Buida III.

Jacob L. Steenwyk is a Howard Hughes Medical Institute awardee of the Life Science Research Foundation at the University of California, Berkeley. Find out more information at his personal website.

Thomas J. Buida III is a senior software and data engineer at Initial State. Find out more information at his personal website.

More Team Members

Nicole King is a Howard Hughes Medical Institute Investigator and Professor of Molecular and Cell Biology at the University of California, Berkeley. Find out more information at her laboratory’s website.

Antonis Rokas is the Cornelius Vanderbilt Chair in Biological Sciences and Director of the Evolutionary Studies Initiative at Vanderbilt University. Research in his laboratory focuses on the study of the DNA record to gain insight into the patterns and processes of evolution. Using a combination of computational and experimental approaches, his lab’s current research aims to understand the molecular foundations of the fungal lifestyle, the reconstruction of the tree of life, and the evolution of human pregnancy. Find out more information at his laboratory’s website.