Improvement of Epitope Prediction Using Peptide Sequence Descriptors and Machine Learning

Abstract

In this work, we improved a previous model used for the prediction of proteomes as new B-cell epitopes in vaccine design. The predicted epitope activity of a queried peptide is based on its sequence, a known reference epitope sequence under specific experimental conditions. The peptide sequences were transformed into molecular descriptors of sequence recurrence networks and were mixed under experimental conditions. The new models were generated using 709,100 instances of pair descriptors for query and reference peptide sequences. Using perturbations of the initial descriptors under sequence or assay conditions, 10 transformed features were used as inputs for seven Machine Learning methods. The best model was obtained with random forest classifiers with an Area Under the Receiver Operating Characteristics (AUROC) of 0.981 ± 0.0005 for the external validation series (five-fold cross-validation). The database included information about 83,683 peptides sequences, 1448 epitope organisms, 323 host organisms, 15 types of in vivo processes, 28 experimental techniques, and 505 adjuvant additives. The current model could improve the in silico predictions of epitopes for vaccine design. The script and results are available as a free repository.

Machine learning-guided discovery and design of non-hemolytic peptides

“…This strategy limits the number of candidates from large peptide libraries by predicting and ranking their biological activities from sequences, so-called Quantitative Structure–Activity/Property Relationship (QSA/PR) studies. Successful QSA/PR applications include the discovery of novel antimicrobial peptides 10 – 13 or epitopes 14 and, the design of anticancer peptides 15 – 17 . Besides to the predictive power of QSA/PR methods, it is essential to consider their limitations, which lie into those of supervised learning.…”

Digital Innovation Enabled Nanomaterial Manufacturing; Machine Learning Strategies and Green Perspectives

“…The sensitivity analysis showed that the reference epitope activity and perturbation of the Shannon entropies were the more important descriptors. The model introduced by the authors is expected to have an impact in the in silico screening of peptides, and contribute to the field of vaccine design by taking advantage of the epitope prediction and the established structure–activity relationships [ 129 ].…”