Strength in numbers: achieving greater accuracy in MHC-I binding prediction by combining the results from multiple prediction tools

Abstract

Background: Peptides derived from endogenous antigens can bind to MHC class I molecules. Those which bind with high affinity can invoke a CD8 + immune response, resulting in the destruction of infected cells. Much work in immunoinformatics has involved the algorithmic prediction of peptide binding affinity to various MHC-I alleles. A number of tools for MHC-I binding prediction have been developed, many of which are available on the web.

Machine learning approaches for epitope prediction

“…This result is consistent with the results reported in the two recent comparison studies (Trost et al, 2007;Lin et al, 2008) where the two servers implementing ARB and SMM (Peters et al, 2006) demonstrated a better performance than Rankpep (Reche et al, 2004) which was implemented using the PSSM method.…”

In silico prediction of Leishmania major‐specific CD8+ epitopes

“…In the end, this would increase the number of predicted epitopes and include peptides with ˃9 amino acid residues [7,8,9] (Data S1). Taken together, mouse-or human-specific epitopes identified in silico hold potential as strong vaccine candidates against cutaneous leishmaniasis.…”

Identification of HLA-A*0201-restricted cytotoxic T lymphocyte epitope from proliferating cell nuclear antigen

“…Moreover at present, we can select peptide that bind to HLA-A*0201 molecule by molecular modeling with accumulation of knowledge on the crystal structures of class I molecules and the sequences of peptides of endogenous origin bound by the class I molecules [19]. A comparative evaluation of prediction serves have been reported [20][21][22]. Many studies show we can achieve higher accuracy by combining the results from different prediction tools.…”