Bootstrapping the Interactome: Unsupervised Identification of Protein Complexes in Yeast

Protein interactions and complexes are major components of biological systems. Recent genome-wide applications of tandem affinity purification (TAP) in yeast have increased significantly the available information on such interactions. From these experiments, protein complexes were predicted with different approaches first from the individual experiments only and later from their combination. The resulting predictions showed surprisingly little agreement and all of the corresponding methods rely on additional training data. In this article, we present an unsupervised algorithm for the identification of protein complexes which is independent of the availability of additional complex information. Based on a bootstrap approach, we calculated intuitive confidence scores for interactions which are more accurate than previous scoring metrics. The complexes determined from this confidence network are of similar quality as the complexes identified by the best supervised approaches. Despite the ...

• Type: Conference
• Year: 2008
• Where: RECOMB
• Authors: Caroline C. Friedel, Jan Krumsiek, Ralf Zimmer
• Keywords: Additional Training Data, Computational Biology, Intuitive Confidence Scores, RECOMB 2008, Tandem Affinity Purification