A fast method to predict protein interaction sites from sequences

A simple method for predicting residues involved In protein interaction sit es is proposed. In the absence of any structural report, the procedure iden tifies linear stretches of sequences as "receptor-binding domains" (RBDs) b y analysing hydrophobicity distribution. The sequences of two databases of non-homologous interaction sites eliciting various biological activities we re tested; 59-80 % were detected as RBDs. A statistical analysis of amino a cid frequencies was carried out in known interaction sites and in predicted RBDs. RBDs were predicted from the 80,000 sequences of the Swissprot datab ase. In both cases, arginine is the most frequently occurring residue. The RBD procedure can also detect residues involved in specific interaction sit es such as the DNA-binding (95 % detected) and Ca-binding domains (83 % det ected). We report two recent analyses; from the prediction of RBDs in seque nces to the experimental demonstration of the functional activities. The ex amples concern a retroviral Gag protein and a penicillin-binding protein. W e support that this method is a quick way to predict protein interaction si tes from sequences and is helpful for guiding experiments such as site-spec ific mutageneses, two-hybrid systems or the synthesis of inhibitors. (C) 20 00 Academic Press.