Abstract: Determining Protein-Protein Interaction (PPI) in biological systems is of significant importance and prediction of PPI has turn out to be a popular research area. Although, different classifiers have been developed for PPI prediction no single classifier seems to be intelligent to predict PPI with high confidence. Here, it is postulated that by combining individual classifiers the accuracy of PPI prediction could be surely improved. In this research, here developed a method called Modified Protein-Protein Interaction Prediction Classifiers Merger (MPPCM) and this method combines output from two PPI prediction tools, GO2PPI and Phyloprof, using Ada Boost algorithm. The performance of MPPCM was tested by Area Under the Curve (AUC) using an assembled gold standard database that contains both positive and negative PPI pairs. Our AUC test showed that MPPCM significantly improved the PPI prediction accuracy over the corresponding individual classifiers. We found that additional classifiers incorporated into MPPCM could lead to further improvement in the PPI prediction accuracy. Furthermore, cross species MPPCM could achieve competitive and even better prediction accuracy compared to the single species MPPCM. This study established a robust pipeline for PPI prediction by integrating multiple classifiers using Ada Boost algorithm. This pipeline will be useful for predicting PPI in nonmodel species.
A. Hepsiba and R. Balasubramanian, 2016. MPPCM: Combing Multiple Classifiers to Improve Protein-Protein Interaction Prediction. Asian Journal of Information Technology, 15: 26-30.