MyProteinNet: build up-to-date protein interaction networks for organisms, tissues and user-defined contexts.
Bottom Line: Users can automatically gather and consolidate data from up to 11 different databases to create a generic protein interaction network (interactome).They can score the interactions based on reliability and filter them by user-defined contexts including molecular expression and protein annotation.MyProteinNet is particularly geared toward building human tissue interactomes, by maintaining tissue expression profiles from multiple resources.
Affiliation: Department of Clinical Biochemistry & Pharmacology, Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel.Show MeSH
Related in: MedlinePlus
Mentions: MyProteinNet uses a Bayesian scoring scheme published previously (22,35). This scheme works in two steps. In the first step, given a specific subset of GO terms (33), e.g. response pathways, the scheme creates positive and negative sets of PPIs as follows. The positive set includes PPIs between proteins annotated to these GO terms and the negative set includes PPIs between proteins that were not annotated to these GO terms. The score of each detection method is then computed based on the ratios of positive to negative PPIs that the detection method identified. In the second step the weight of each PPI is calculated according to the detection methods that identified it using a Bayesian computation (21). Since some detection methods identify only few PPIs, these methods and the corresponding PPIs cannot be scored reliably. To overcome this limitation, we manually combined detection methods that differ only slightly. For example, anti bait coimmunopercipitation (MI:0006) and anti tag coimmunopercipitation (MI:0007) were both grouped as coimmunopercipitation methods. MyProteinNet thus scores 37 combined detection methods (Supplementary Table S2). A diagram describing the distribution of PPI weights is provided as part of the output (Figure 1).
Affiliation: Department of Clinical Biochemistry & Pharmacology, Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel.