A Bayesian approach for estimating protein–protein interactions by integrating structural and non-structural biological data

摘要

Accurate elucidation of genome wide protein–protein interactions is crucial for understanding thernregulatory processes of the cell. High-throughput techniques, such as the yeast-2-hybrid (Y2H) assay,rnco-immunoprecipitation (co-IP), mass spectrometric (MS) protein complex identification, affinityrnpurification (AP) etc., are generally relied upon to determine protein interactions. Unfortunately, eachrntype of method is inherently subject to different types of noise and results in false positive interactions.rnOn the other hand, precise understanding of proteins, especially knowledge of their functionalrnassociations is necessary for understanding how complex molecular machines function. To solve thisrnproblem, computational techniques are generally relied upon to precisely predict protein interactions.rnIn this work, we present a novel method that combines structural and non-structural biological data tornprecisely predict protein interactions. The conceptual novelty of our approach lies in identifying andrnprecisely associating biological information that provides substantial interaction clues. Our modelrncombines structural and non-structural information using Bayesian statistics to calculate the likelihoodrnof each interaction. The proposed model is tested on Saccharomyces cerevisiae’s interactions extractedrnfrom the DIP and IntAct databases and provides substantial improvements in terms of accuracy, precision,rnrecall and F1 score, as compared with the most widely used related state-of-the-art techniques.