Opportunities for machine learning to predict cross-neutralization in FMDV serotype O

Abstract

Accurately estimating cross-neutralization between serotype O foot-and-mouth disease viruses (FMDVs) is critical for guiding vaccine selection and disease management. In this study, we developed a machine learning approach to estimate r1 values—an established measure of antigenic similarity—using VP1 sequence data and published virus neutralization titer (VNT) results. Our dataset comprised 108 serum-virus pairs representing 73 distinct FMDV strains. We applied Boruta feature selection and random forest classifiers, optimizing model performance through tenfold cross-validation and sub-sampling to address class imbalance. Predictors included pairwise amino acid distances, site-specific polymorphisms, and differences in potential N-glycosylation sites. Using a 0.3 r1 threshold to define cross-neutralization, the final model achieved high accuracy (0.96), sensitivity (0.93), and specificity (0.96) in training, and performed robustly on independent test sets - accuracy was 0.75 (95% CI 0.60 and 0.90), F1 score 0.86% and PPV 0.77. Importantly, key VP1 residues—positions 48, 100, 135, 150, and 151—emerged as strong predictors of antigenic relationships. Our results demonstrate the utility of integrating routinely generated genomic data with machine learning to inform vaccine candidate selection and anticipate immune interactions among circulating FMDV strains. This approach offers a practical tool for accelerating vaccine decision-making and can be adapted to other FMDV serotypes. The latest version of the r1 predictive model is available for access via a Shiny dashboard (https://dmakau.shinyapps.io/PredImmune-FMD/). Author summary: Foot-and-mouth disease (FMD) is a highly contagious virus that threatens the health of livestock and the livelihoods of farmers around the world. Controlling outbreaks depends on using vaccines that protect against the specific virus strains circulating in the field. However, testing whether a vaccine will be effective typically requires complex laboratory experiments that are not always possible during urgent outbreaks. In this study, we developed a computer-based method using machine learning to predict how closely a vaccine strain matches a field virus by comparing their genetic sequences, focusing on a key virus protein involved in immunity. We collected data from 108 tests reported in previous studies and trained a model that accurately—over 90% of the time—identified vaccine and virus strain combinations likely to provide good protection. This approach can speed up vaccine selection, help prepare for outbreaks faster, and reduce reliance on costly lab tests, particularly in regions with limited resources. Importantly, we have also translated our model into an interactive, user-friendly tool called PredImmune-FMD available at https://dmakau.shinyapps.io/PredImmune-FMD/, enabling researchers, field veterinarians, and decision-makers to explore predictions in real time.

Publication information

Originally published in PLOS Computational Biology on 2025-09-17. Original DOI: 10.1371/journal.pcbi.1013491.

The original full text is available here under CC0 1.0. The displayed publication date is the original publication date.