[Accompanying Dataset for PHIStruct] ColabFold-Predicted Structures of Receptor-Binding Proteins

This dataset contains protein structures, computationally predicted via ColabFold, of 19,081 non-redundant (i.e., with duplicates removed) receptor-binding proteins from 8,525 phages across 238 host genera. We identified these receptor-binding proteins based on GenBank annotations. For phage sequences without GenBank annotations, we employed a pipeline that uses the viral protein library PHROG and the machine learning model PhageRBPdetect. 

More details can be found in our paper "PHIStruct: Improving phage-host interaction prediction at low sequence similarity settings using structure-aware protein embeddings." The project page is https://github.com/bioinfodlsu/PHIStruct. Our paper is published in Bioinformatics: https://doi.org/10.1093/bioinformatics/btaf016

Our research was supported with Cloud TPUs from Google's TPU Research Cloud (TRC) and with computing resources from the Machine Learning eResearch Platform (MLeRP) of Monash University, University of Queensland, and Queensland Cyber Infrastructure Foundation Ltd.