Baylor College of Medicine

Lichtarge Computational Biology Lab

About Our Research

Our lab marries computation with experiments to study the molecular evolution of genes and pathways. We seek broadly to characterize biological mechanisms and functions, and how they may be corrupted by genetic mistakes or re-engineered to new purpose. The long-term goals are to design new personalized therapies and to harness the synthetic potential of organisms. Shorter term goals are to interpret the action of human genome variations on health and disease.

Our algorithms broadly merge mathematical and evolutionary principles together with machine learning and artificial intelligence. As a result, they enable multi-scale data integration and precise control of molecular functions. This has led to discoveries in diverse systems, from E. Coli to Humans, including in drug resistance, G protein signaling, malaria and cancer. Starting from structural bioinformatics, newer interests now include network theory, text-mining and cognitive computing. Technically, we draw upon a wide range of disciplines to address fundamental questions in structural biology, clinical genomics and precision medicine.

Specific examples includes a network compression scheme that made tractable the diffusion of information across nearly 400 species. This approach uncovered a possible mechanism for the best current drug against malaria. Other network studies, reasoned over the entire PubMed literature to discover new kinases and protein interactions for p53. A distinct line of research quantifies the evolutionary action (EA) of mutations on fitness and bridges between molecular and population genetics. EA correlates with experimental loss of function in proteins; with morbidity and mortality in people; and with purifying gene selection in population. In some head and neck cancer patients, it stratifies outcomes and suggests alternate therapy for some patients. In the coming years we hope to unite these different approaches into a coherent path to precision therapy personalized to patients on a case by case basis.

Our Research

The interactions of proteins amongst themselves and with other macromolecules control cellular networks and govern life. In order to decipher these interactions and to identify specific and appropriate targets for drugs, our work aims to identify the functional sites that mediate protein binding and to predict how their component amino acids contribute to functional specificity.

Our approach is based on the Evolutionary Trace (ET, Lichtarge 1996, Mihalek 2004). This method entails ranking the relative functional importance of amino acids in a protein sequence by correlating their variations during evolution with divergences in the phylogenetic tree of that sequence family. This approach is similar to laboratory-based mutational scanning, but it exploits the vast number of mutations and assays that were already tested through evolution, and that are increasingly retrievable from sequence and structure databases.

Press Release


Recent Lab News

Computer says “try this”

APPLES, mushrooms and pork sounds a promising recipe for a kebab, but the average barbecuer might balk at adding strawberries. According to John Gordon of IBM, however, the result is ...

Read more
Compressed interaction networks assign gene function

Assigning function to genes remains a big challenge in biology. Researchers often turn to gene or protein networks to infer the role of a particular gene. Local network analysis identifies ...

Read more
New Watson incarnation crunches academic literature

IBM has unveiled a new application for Watson — the Jeopardy winning super-computing system — designed to help academic researchers rapidly process scientific literature. During a Thursday event at New ...

Read more
"Jeopardy!"-winning computer now using its brain for science

Watch out, Sherlock, there's a new Dr. Watson in town. IBM's Watson, the computer that famously won the quiz show 'Jeopardy!', is now helping researchers make scientific discoveries. The artificially ...

Read more
Evolutionary Trace

The structure and function of proteins underlie most aspects of biology and their mutational perturbations often cause disease. A central question is which protein residue positions are important and cluster to form functional sites.

As a solution to this question, the Evolutionary Trace (ET) computes the relative rank of functional and structural importance among protein homologs sequence positions. The rank is lower if sequence positions vary among evolutionarily closer homologs and higher if the positions vary among evolutionarily distant homologs. Thus, ET uses evolutionary distances as a proxy for functional distances to correlate genotype variations with phenotype, or fitness, variations. This approach identifies functional determinants, predicts function, guides the mutational redesign of functional and allosteric specificity, and interprets the action of coding sequence variations in proteins, people, and populations.