Analyzing Protein Structure-Function Correlations Using Statistical Geometry
Majid Masso1, Losif Vaisman2
1mmasso@gmu.edu, George Mason University; 2ivaisman@gmu.edu, George Mason University
A novel technique employing computational geometry, namely Delaunay tessellation of protein structure represented by alpha-carbons in 3D space, yields an objective and robust definition of four nearest-neighbor residues as well as a 4-body statistical potential function. Using this approach, an isolated 99 amino acid subunit of the HIV-1 protease homodimer is analyzed, yielding individual residue environment scores as well as a total topological score for the monomer. A comparison with the identical chain in a dimeric state with an inhibitor elucidates changes to structural stability. Predictably, protease residues shown to undergo the greatest impact are those forming the dimer interface and flap region, as well as those known to be involved in inhibitor binding. Additionally, topological scores for all 1881 single point mutants of the HIV-1 protease are evaluated, and a comprehensive mutational profile is developed. Of these mutants, 536 are compared with their experimentally synthesized counterparts. The analysis reveals a clear relationship between mutant structure (mean change in the topological score of the computational mutants from wild-type) and function (observed phenotype of the experimental mutants).