New demands and expectations for predicting protein-protein interactions have been emerging with the development of computer power and experimental techniques.Meeting these needs,the role of computational tools like protein-protein docking has significantly extended to accommodate the newly-occurring challenges like oligomer modeling of monomers,water prediction,and protein-peptide binding.In addition,with the rapid development of structural proteomics projects,more and more structures of protein complexes are being experimentally determined.Therefore,how to efficiently incorporate the vast amount of biological information in traditional protein-protein docking has been a challenge in the protein docking field.To address these challenges,we have presented a hybrid method to combine the advantages of template-based and template-free docking by integrating them together.The basic procedure is to first construct the complex of individual structures based on template-based docking if a template is available and then use the component structures of the complex as input for a free protein-protein docking to increase sampling of putative binding modes.We have participated all the recent CPARI experiments (http://www.ebi.ac.uk/msd-srv/capri/) with our hybrid docking protocol.It was shown that our hybrid protocol was one of the top algorithms for the CAPRI-CASP 11 challenge of oligomer modeling and was able to predict correct models for 16 out 25 targets,while also made correct predictions on other CAPRI challenges like protein-peptide binding.