Epitopedia: identifying molecular mimicry of known immune epitopes
-
Updated
Feb 18, 2022 - JavaScript
Epitopedia: identifying molecular mimicry of known immune epitopes
HLA-A*02:01 nine-mer peptide binding prediction from sequence, with grouped validation and fully traceable metrics.
Standalone Python package for annotating protein sequences with biological and physicochemical properties (iFeature, IEDB, SignalP) for reverse vaccinology. Extracted from VacSol-ML(ESKAPE); supports multi-allele MHC-I/II epitope prediction. Provides both a Python API and a CLI
Comparative evaluation of Machine Learning and Deep Learning models for peptide immunogenicity prediction using physicochemical feature engineering and class imbalance handling.
Chapter 6 of Machine Learning for Biology. Sequence-only immunogenicity prediction on 1,200 HLA-A*02:01 IEDB binders, cluster-split and pre-registered. Network 0.610 AUPRC, binding-only baseline 0.591, chance 0.485.
To associate your repository with the iedb topic, visit your repo's landing page and select "manage topics."