James Thompson

Previous Degrees

M.S. Bioinformatics, Rochester Institute of Technology.
B.S. Bioinformatics, Rochester Institute of Technology.

Research

My research is focused on the improvement of structure prediction for proteins. It has been known for over 40 years that the primary amino acid sequence of a protein contains all of the information necessary to predict the protein's fully folded native conformation. Based on this idea, protein structure prediction has been formulated as a search problem in which evaluation by physically-motivated energy functions can discriminate between native and non-native conformations for a given protein sequences.

Given the high dimensionality and vast size of the solution space to this problem, many successful approaches to protein structure prediction have identified promising candidate structures by leveraging knowledge from known protein sequences to create new candidate structures, including Rosetta, a software program developed within my lab. I am currently developing methods to better use information from the protein data bank to improve the prediction of protein structures using Rosetta.

Bibliography

Thompson, J. and S. Gopal (2006). "Genetic algorithm learning as a robust approach to RNA editing site prediction." BMC Informatics 7: 145.

Das, R., R. Vernon, et al. (2006). Protein structure prediction by free modeling and Rosetta@home in CASP7. 7th Community Wide Experiment on the Critical Assessment of Techniques for Protein Structure Prediction. Asilomar Conference Center, Pacific Grove, CA.

Thompson, J. (2006). Ab-initio structure prediction in CASP7. UW Genome Sciences Research Report. Seattle, WA.
Das, R., B. Qian, et al. (2007). "Structure prediction for CASP7 targets using extensive all-atom refinement with Rosetta@home " Proteins: Structure, Function and Bioinformatics 69(S8): 118-128.