AXEL-F

AXEL-F integrates peptide–MHC binding affinities with source-protein mRNA expression to estimate the likelihood of peptide presentation by MHC class I for T cell epitope and neoantigen prediction.

Key Features:

• Integration of peptide-MHC binding and antigen expression: AXEL-F combines peptide–MHC binding affinity data with RNA expression levels of source proteins using a biophysically motivated likelihood function to estimate presentation probability.
• Improved prediction accuracy: By jointly modeling binding and expression, AXEL-F better discriminates eluted ligands from background peptides and improves neoantigen prediction for T cell recognition.
• Use of RNA-Seq and TCGA expression data: AXEL-F can incorporate patient-specific RNA-Seq expression or, when unavailable, use cancer-type matched expression from The Cancer Genome Atlas (TCGA) to estimate gene expression.

Scientific Applications:

• Neoantigen prioritization for cancer immunotherapy: AXEL-F refines selection of MHC class I-presented neoantigens for personalized cancer vaccines and T cell–based therapies.
• Epitope identification and eluted ligand discrimination: AXEL-F aids identification of T cell-recognized epitopes by distinguishing naturally eluted ligands from random peptide backgrounds.

Methodology:

AXEL-F computationally combines peptide–MHC binding affinities with RNA expression levels via a biophysically motivated likelihood function and accepts patient RNA-Seq or TCGA cancer-type matched expression data.