Characterization of protein interactions

Abstract

Available information on molecular interactions between proteins is currently incomplete with regard to detail and comprehensiveness. Although a number of repositories are already devoted to capture interaction data, only a small subset of the currently known interactions can be obtained that way. Besides further experiments, knowledge on interactions can only be complemented by applying text extraction methods to the literature. Currently, information to further characterize individual interactions can not be provided by interaction extraction approaches and is virtually nonexistent in repositories. We present an approach to not only confirm extracted interactions but also to characterize interactions with regard to four attributes such as activation vs. inhibition and protein-protein vs. protein-gene interactions. Here, training corpora with positional annotation of interacting proteins are required. As suitable corpora are rare, we propose an extensible curation protocol to conveniently characterize interactions by manual annotation of sentences so that machine learning approaches can be applied subsequently. We derived a training set by manually reading and annotating 269 sentences for 1090 candidate interactions; 439 of these are valid interactions, predicted via support vector machines at a precision of 83% and a recall of 87%. The prediction of interaction attributes from individual sentences on average yielded a precision of about 85% and a recall of 73%.