Large-scale Protein-Protein Interaction prediction using novel kernel methods

Knowledge of Protein-Protein Interactions (PPIs) call give its new insights molecular mechanisms and properties of' the cell. In this paper, xve propose a novel clonia i ii- based kernel niethod to predict PPIs. A new kernel that ITICaSLINS the similarity betwecil protein pairs based oil a new leatUrC representation is developed and applied to a large scale PPI database. Fxperinicrital rmlltS demonstrate its effectivericss. Futhermore, we CValLlatC the problein of' cross-spccies PPI prediction and the effect of the number of negative samples on the perfonnance of PPI predictions, which are two fundamental problems in most in silico PPI methods.