HotSprint: Database of Computational Hot Spots at Protein Interfaces E. Guney, N. Tuncbag, A. Gursoy, O. Keskin Center for Computational Biology and Bioinformatics, Koc University, Rumeli Feneri Yolu, Sariyer, 34450, Istanbul We present a new database of computational hot spots at protein interfaces: HotSprint. Sequence conservation and solvent accessibility of interface residues are used to detect the computational hot spots. Hot spots are residues comprising only a small fraction of interfaces yet accounting for the majority of the binding energy. HotSprint contains data for approximately 32000 protein interfaces among 49512 protein interfaces extracted from the multi-chain structures in Protein Data Bank (PDB) as of February, 2006. The conserved residues on the interfaces with certain buried ASA and complex ASA thresholds are flagged as computational hot spots. The predicted hot spots are observed to correlate with the experimental hot spots with an accuracy of 72%. Several machine learning methods (SVM, Decision Trees and Decision Lists) are also applied to predict the hot spots, results reveal that our empirical approach is slightly better. A database web site, HotSprint allows users to browse and query the hot spots in protein interfaces. Evolutionary conservation, solvent accessibility in complex and monomer forms can be fetched using PDB or interface

identifier.

HotSprint

is

available

at

http://prism.ccbb.ku.edu.tr/hotsprint; and it provides information on interface residues that are functionally and structurally important as well as the evolutionary history and solvent accessibility of the residues on the interfaces.