The task of predicting the interface of a given protein using only the structure of the unbound protein, is an important goal. Many groups have attempted tackling this problem from different fronts and using different approaches. We preset here 10 popular protein-protein interface prediction servers. This might prove especially helpful to those who take part in the ongoing CAPRI round.
And the servers are (in random order):
- PPI-pred - Predicts protein-protein binding sites using a combination of surface patch analysis and a support vector machine (SVM) trained on 180 proteins involved in both obligate and non-obligate interactions. Cite: Improved prediction of protein-protein binding sites using a support vector machines approach.
- meta-PPISP - Built on three individual web servers: cons-PPISP, PINUP, and Promate. A linear regression method, using the raw scores of the three severs as input, was trained on a set of 35 nohomologous proteins. Cite: meta-PPISP: a meta web server for protein-protein interaction site prediction.
- cons-PPISP - A consensus neural network method for predicting protein-protein interaction sites. Given the structure of a protein, cons-PPISP will predict the residues that will likely form the binding site for another protein. The inputs to the neural network include position-specific sequence profiles and solvent accessibilities of each residue and its spatial neighbors. The neural network is trained on known structures of protein-protein complexes on a benchmark set of 22 protein complexes. Cite: Prediction of interface residues in protein-protein complexes by a consensus neural network method: Test against NMR data.
- PINUP - Trained on a 57-protein data set and employs: (i) effective residue-energy score and accessible-surface-area-dependent interface-propensity, (ii) isolation of functional constraints contained in the conservation score from the structural constraints through the combination of residue-energy score (for structural constraints) and conservation score and (iii) a consensus region built on top-ranked initial patches. Cite: Protein binding site prediction using an empirical scoring function.
- ProMate - A set of surface dots at a constant density is extracted for the query protein. These dots serve as centers for 10A-radius circles over the protein’s surface. Using the distributions of the properties, that have been found to distinguish binding from non-binding surfaces, the predictor evaluates the probability of each circle to appear at the interface. Cite: ProMate: a structure based prediction program to identify the location of protein-protein binding sites.
- SPPIDER - A representation which integrates enhanced relative solvent accessibility (RSA) predictions with high resolution structural data. RSA prediction-based fingerprints of protein interactions significantly improve the discrimination between interacting and noninteracting sites. Cite: Prediction-based Fingerprints of Protein-Protein Interactions.
- WHISCY - primarily based on conservation, but it also takes into account structural information. A sequence alignment is used to calculate a prediction score for each surface residue of your protein. Cite: WHISCY: What information does surface conservation yield? Application to data-driven docking.
- ConSurf - A useful and user-friendly tool that enables the identification of functionally important regions on the surface of a protein or domain. Based on the phylogenetic relations between its close sequence homologues. Cite: ConSurf: the projection of evolutionary conservation scores of residues on protein structures
- InterProSurf - Predicts interacting amino acid residues in proteins that are most likely to interact with other proteins. The prediction method is based on solvent accessible surface area of residues in the isolated subunits, a propensity scale for interface residues and a clustering algorithm to identify surface regions with residues of high interface propensities. Cite: InterProSurf: a web server for predicting interacting sites on protein surfaces.
- ProteMot - The Protemot (Protein motif) web server predicts protein binding sites based on the interaction templates automatically extracted from the compound crystals in the Protein Data Bank (PDB). Cite: Protemot: prediction of protein binding sites with automatically extracted geometrical templates
Zhou & Qin wrote a very nice review of interface prediction methods and the evaluation of such methods and compared the performance of 6 of the servers presented above. (Interaction-site prediction for protein complexes: a critical assessment) according to their assessment on two data sets the best performing server, out of those 6, is meta-PPSIP (also by Zhou & Qin).
Good luck with your predictions! Know of any good server that is not on the list ? Tell us in the comments.
RECENT COMMENT