epitope1D: accurate taxonomy-aware B-cell linear epitope prediction.

Area of Science:

• Immunoinformatics
• Computational Biology
• Machine Learning

Background:

• Identifying B-cell epitopes is crucial for vaccine design, diagnostics, and antibody production.
• Existing computational methods for epitope prediction have limitations in performance and interpretability, hindering biological insights.

Purpose of the Study:

• To develop an explainable machine learning method for accurate identification of linear B-cell epitopes.
• To improve predictive performance and provide biological insights into epitope identification.

Main Methods:

• Developed epitope1D, an explainable AI model utilizing novel graph-based sequence descriptors and Organism Ontology information.
• Employed Cutoff Scanning Matrix algorithm for feature representation.
• Evaluated performance using cross-validation and blind tests on benchmark datasets.

Main Results:

• Achieved an Area Under the ROC curve of up to 0.935, demonstrating robust predictive performance.
• epitope1D outperformed state-of-the-art computational tools in comprehensive comparisons.
• The model allows for the combination and interpretation of biologically meaningful features.

Conclusions:

• epitope1D offers a significant advancement in predicting linear B-cell epitopes with high accuracy and interpretability.
• The tool enhances biological understanding in vaccine design and related fields.
• A user-friendly web server and API are available for broader accessibility.