Discovery of protein complexes with core-attachment structures from Tandem Affinity Purification (TAP) data

Area of Science:

• Proteomics
• Computational Biology
• Systems Biology

Background:

• Cellular functions rely on protein complexes formed by interacting proteins.
• Tandem Affinity Purification (TAP) is a key method for identifying protein interactions.
• Current computational methods for protein complex prediction from TAP data lose information by converting to pairwise interactions.

Purpose of the Study:

• To develop a novel method, CACHET, for detecting protein complexes with core-attachment structures directly from TAP data.
• To overcome limitations of existing methods that convert complex TAP data into pairwise protein-protein interaction networks.
• To leverage the inherent structure of TAP data for more accurate protein complex identification.

Main Methods:

• CACHET models TAP data as a bipartite graph (baits and preys).
• It identifies protein-complex cores using reliable, non-redundant bicliques.
• Attachment proteins are incorporated into the identified cores to form complete complexes.

Main Results:

• CACHET demonstrated superior prediction accuracy (F-measure, functional homogeneity) on large-scale TAP datasets.
• The method successfully identified protein complexes with distinct core-attachment structures.
• Predicted structures offer valuable biological insights into protein complex organization.

Conclusions:

• CACHET provides a more accurate and informative approach to protein complex prediction from TAP data.
• The core-attachment structures identified by CACHET enhance understanding of protein complex functional organization.
• This method offers significant advantages over traditional protein-protein interaction network-based approaches.