Integrative approach for computationally inferring protein domain interactions.

Area of Science:

• Computational Biology
• Bioinformatics
• Systems Biology

Background:

• High-throughput experimental methods like yeast-two-hybrids generate large volumes of protein interaction data.
• Experimental data can be erroneous and often lack sufficient functional information.
• There is a need for in silico approaches to validate and annotate protein interactions.

Purpose of the Study:

• To develop an integrative computational approach for deriving domain-domain interactions.
• To utilize derived domain interactions for validating, annotating, and predicting protein-protein interactions.

Main Methods:

• An integrative computational approach was developed to derive domain-domain interactions.
• Multiple data sources were used, including protein interactions, protein complexes, and Rosetta Stone sequences.
• The derived domain interactions were applied to predict and validate protein-protein interactions.

Main Results:

• A method was successfully developed to computationally derive putative domain-domain interactions.
• The derived domain interactions proved useful in predicting and validating protein-protein interactions.
• A database of these putative domain interactions is now available.

Conclusions:

• Computational derivation of domain-domain interactions is a valuable approach for biological data analysis.
• This method enhances the validation, annotation, and prediction of protein-protein interactions.
• The publicly available database facilitates further research in protein interaction networks.