PHIStruct: Improving phage-host interaction prediction at low sequence similarity settings using structure-aware protein embeddings

Mark Edward M Gonzales; Jennifer C Ureta; Anish M S Shrestha

doi:10.1093/bioinformatics/btaf016

PHIStruct: Improving phage-host interaction prediction at low sequence similarity settings using structure-aware protein embeddings

Bioinformatics. 2025 Jan 13:btaf016. doi: 10.1093/bioinformatics/btaf016. Online ahead of print.

Authors

Mark Edward M Gonzales^{1

2}, Jennifer C Ureta^{1

2

3}, Anish M S Shrestha^{1

2}

Affiliations

¹ Bioinformatics Lab, Advanced Research Institute for Informatics, Computing and Networking, De La Salle University, Manila, 1004, Philippines.
² College of Computer Studies, De La Salle University, Manila, 1004, Philippines.
³ Walter and Eliza Hall Institute of Medical Research, Melbourne, Victoria 3052, Australia.

PMID: 39804673
DOI: 10.1093/bioinformatics/btaf016

Abstract

Motivation: Recent computational approaches for predicting phage-host interaction have explored the use of sequence-only protein language models to produce embeddings of phage proteins without manual feature engineering. However, these embeddings do not directly capture protein structure information and structure-informed signals related to host specificity.

Results: We present PHIStruct, a multilayer perceptron that takes in structure-aware embeddings of receptor-binding proteins, generated via the structure-aware protein language model SaProt, and then predicts the host from among the ESKAPEE genera. Compared against recent tools, PHIStruct exhibits the best balance of precision and recall, with the highest and most stable F1 score across a wide range of confidence thresholds and sequence similarity settings. The margin in performance is most pronounced when the sequence similarity between the training and test sets drops below 40%, wherein, at a relatively high-confidence threshold of above 50%, PHIStruct presents a 7% to 9% increase in class-averaged F1 over machine learning tools that do not directly incorporate structure information, as well as a 5% to 6% increase over BLASTp.

Availability and implementation: The data and source code for our experiments and analyses are available at https://github.com/bioinfodlsu/PHIStruct.

Supplementary information: Supplementary data are available at Bioinformatics online.

Keywords: deep learning; phage-host interaction prediction; protein language models; protein structure; receptor-binding proteins; representation learning.