Send to

Choose Destination
Proc Natl Acad Sci U S A. 2013 Jun 25;110(26):10872-7. doi: 10.1073/pnas.1308621110. Epub 2013 Jun 11.

Optimization criteria and biological process enrichment in homologous multiprotein modules.

Author information

Center for Computational Biology and Division of Computer Science, University of California, Berkeley, CA 94720, USA.


Biological process enrichment is a widely used metric for evaluating the quality of multiprotein modules. In this study, we examine possible optimization criteria for detecting homologous multiprotein modules and quantify their effects on biological process enrichment. We find that modularity, linear density, and module size are the most important criteria considered, complementary to each other, and that graph theoretical attributes account for 36% of the variance in biological process enrichment. Variations in protein interaction similarity within module pairs have only minor effects on biological process enrichment. As random modules increase in size, both biological process enrichment and modularity tend to improve, although modularity does not show this upward trend in modules with size at most 50 proteins. To adjust for these trends, we recommend a size correction based on random sampling of modules when using biological process enrichment or other attributes to evaluate module boundaries. Characteristics of homologous multiprotein modules optimized for each of the optimization criteria are examined.

[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for HighWire Icon for PubMed Central
Loading ...
Support Center