Discovering Reliable Protein Interactions from High-Throughput Experimental Data using Network Topology

Current protein-protein interaction(PPI)detection via high-throughput experimental methods such asyeast-two-hybrid has been reported to be highly erroneous, leadingto potentially costly spurious discoveries. This work introduces anovel measure called IRAP, i.e., ''Interaction Reliabilityby Alternative Path'', for assessing the reliability of proteininteractions based on the underlying topology of the PPI network.

A candidate PPI is consideredto be reliable if it is involved in a closed loop in which thealternative path of interactions between the two interactingproteins is strong. We devise an algorithm calledAlternativePathFinder to compute the IRAP value for eachinteraction in a complex PPI network. Validation of the IRAP as ameasure for assessing the reliability of PPIs is performed withextensive experiments on yeast PPI data.

Positive experimental resultsdemonstrate that IRAP is a good measure for assessing thereliability of PPIs from conventional high-throughput experiments.

---by Jin Chen, Wynne Hsu, Mong Li Lee and See-Kiong Ng, Artifical Intelligence in Medicine, 2005