v.s. IG1/2 
v.s. PathRatio 



Experimental limitations in high-throughput proteinprotein interaction detection methods have resulted in low quality interaction datasets containing sizable fractions of false positives and false negatives. Small-scale, focused experiments are needed to complement the high throughput methods to extract true protein interactions. However, the naturally vast interactomes would require much more scalable approaches.


  • We describe a novel method called IRAP* for the computational repurification of the highly erroneous experimentally-derived protein interactomes. Our method involves an iterative process of removing interactions that are confidently identified as false positives and adding interactions detected as false negatives into the interactomes. Identification of both false positives and false negatives are performed using interaction confidence metrics computed with the topological information of the underlying interaction network. Potential false positives are identified amongst the detected interactions as those with very low computed reliability values, while potential false negatives are discovered as the undetected interactions with high computed reliability values.
  • Our results from applying IRAP* on large-scale interaction data sets generated by the popular yeast-two-hybrid assays for yeast, fruit fly and worm show that the computationally repurified interaction data sets contain potentially lower fractions of false positive and false negative errors based on functional homogeneity.

Paper and supplementary document

The paper "Increasing confidence of protein interactomes using network topological metrics" will appear soon. The supplementary document is availabel here.


Prof. Wynne Hsu
Department of Computer Science,
School of Computing,
National University of Singapore,
Singapore 117590.

Dr. See-Kiong Ng
Knowledge Discovery Department,
Institute for Infocomm Research,
Singapore 119613



All rights reserved.