Prediction of missing common genes for disease pairs using network based module separation on incomplete human interactome

BACKGROUND: Identification of common genes associated with comorbid diseases can be critical in understanding their pathobiological mechanism. This work presents a novel method to predict missing common genes associated with a disease pair. Searching for missing common genes is formulated as an optimization problem to minimize network based module separation from two subgraphs produced by mapping genes associated with disease onto the interactome. RESULTS: Using cross validation on more than 600 disease pairs, our method achieves significantly higher average receiver operating characteristic ROC Score of 0.95 compared to a baseline ROC score 0.60 using randomized data. CONCLUSION: Missing common genes prediction is aimed to complete gene set associated with comorbid disease for better understanding of biological intervention. It will also be useful for gene targeted therapeutics related to comorbid diseases. This method can be further considered for prediction of missing edges to complete the subgraph associated with disease pair.