Statistical coevolution analysis and molecular dynamics: identification of amino acid pairs essential for catalysis.

Molecular dynamics (MD) simulations of HhaI DNA methyltransferase and statistical coupling analysis (SCA) data on the DNA cytosine methyltransferase family were combined to identify residues that are coupled by coevolution and motion. The highest ranking correlated pairs from the data matrix product (SCA.MD) are colocalized and form stabilizing interactions; the anticorrelated pairs are separated on average by 30 A and form a clear focal point centered near the active site. We suggest that these distal anti-correlated pairs are involved in mediating active-site compressions that may be important for catalysis. Mutants that disrupt the implicated interactions support the validity of our combined SCA.MD approach.