SeRenDIP — SEquence-based Random forest predictor with lENgth and Dynamics for Interacting Proteins

Please do not link this web address!
For stable links, please refer to www.ibi.vu.nl/programs/ or www.ibi.vu.nl/programs/serendipwww/index.php.

The SeRenDIP server offers a simple interface to our random-forest based methods for predicting protein-protein interaction (PPI) positions from a single input sequence. We have published predictors for heteromeric PPIs ('Heteromeric') and for generic homomeric/​heteromeric PPIs ('Combined'). We have also added two development predictors for Epitope-Antibody interactions, specific for epitopes ('Epitope'), or generic for either epitope or heteromeric interactions ('EpiComb'); the manuscript for the epitope interaction prediction is in preparation. You can input your sequence of interest, and obtain a table of predicted interface positions. Please note that runtimes are typically around or below three hours, but may be up to 15 hours, depending on the number of blast hits for your query. If you need more performance, you may instead rather want to use the stand alone version supplied on the download page.	Paste in your input sequence:   Showcase 1YVB:A or  upload your alignment: Select random forest model trained on Combined Heteromeric EpiComb Epitope dataset.
Example output: pdb 1YVB chain A Plasmodium falciparum Cysteine Protease Please cite: Qingzhen Hou1, Bas Stringer, Katharina Waury, Reza Haydarlou, Sanne Abeln, Jaap Heringa, K. Anton Feenstra SeRenDEP: SEquence-based Random forestpredictor with lENgth and Dynamics for Epitope Prediction. (in preparation) 1Department of Biostatistics, School of Public Health & National institute of health data science of China; Shandong 250002, P. R. China Qingzhen Hou1, Paul De Geest, Christian Griffioen, Sanne Abeln, Jaap Heringa, K. Anton Feenstra. SeRenDIP: SEquential REmasteriNg to DerIve profiles for fast and accurate predictions of PPI interface positions. Bioinformatics 35, pp 4794–4796, 2019, doi: 10.1093/bioinformatics/btz428 . 13BIO-BioInfo – BioModeling, BioInformatics & BioProcesses, Université Libre de Bruxelles Qingzhen Hou, Paul De Geest, Wim F. Vranken1, Jaap Heringa and K. Anton Feenstra. Seeing the Trees through the Forest: Sequence-based Homo- and Heteromeric Protein-protein Interaction sites prediction using Random Forest. Bioinformatics 33 pp 1479–1487, 2017, doi: 10.1093/bioinformatics/btx005 1Interuniversity Institute of Bioinformatics in Brussels, ULB-VUB & Structural Biology Brussels, VUB & Structural Biology Research Centre, VIB; Brussels. External prediction methods used: DynaMine: Cilia, E., Pancsa, R., Tompa, P., Lenaerts, T., and Vranken, W. (2013). From protein sequence to dynamics and disorder with DynaMine. Nature communications, 4:2741. Cilia, E., Pancsa, R., Tompa, P., Lenaerts, T., and Vranken, W. F. (2014). The DynaMine web-server: predicting protein dynamics from sequence. Nucleic acids research, 42(W1):W264—W270. NetSurfP: Petersen, B., Petersen, T. N., Andersen, P., Nielsen, M., and Lundegaard, C. (2009). A generic method for assignment of reliability scores applied to solvent accessibility predictions. BMC structural biology, 9(1):1.
The SeRenDIP web server developed and Copyright (c) by K. Anton Feenstra, Paul de Geest and Qingzhen Hou.