The binding of antigenic peptides to the major histocompatibility complex (MHC) is a pre-requisite for T cell recognition, which in turn can lead to an immune response. The reliable prediction of the affinity of candidate peptides for MHC is important for determining their potential antigenicity, and thus influences decisions on their inclusion in neoantigen-based cancer vaccines and other medical applications. Here, we present a rapid and reliable computational approach that combines a popular, sequence-based artificial neural network method, NetMHCpan 4.0, with 3D structural modeling to predict the affinity of peptides to MHC. The structural modeling involves docking candidate peptides to the MHC using the programs MODELLER and Rosetta FlexPepDock. We find that the ensembles of bound peptide conformations thus generated are less variable in geometry for strong binders than for low-affinity peptides. We tested our approach on 1,271 peptide sequences for which the experimental dissociation constants for the peptide-MHC complexes associated with the murine MHC allele, H-2Db are known. Using our structure-based approach in a stand-alone manner drastically increases the statistical specificity, reducing the number of false positives in the non-binding affinity range (kD &gt; 6,500 nM) to less than 2%, compared to 12% with the sequence-based method. Furthermore, filtering candidates generated with NetMHCpan 4.0 with the structure-based predictor led to an increase in the positive predictive value (PPV) of the peptides correctly predicted to bind very strongly (i.e., kD &lt;100 nM) from 40 to 52% (p=0.027). The combined method may be useful in various therapeutic applications, including the rapid design of effective peptide-based vaccines.

Combining rapid 3D structural modeling with machine learning to increase specificity in peptide-based vaccine design

Macromolecular Chemistry: The Second Century

[Author Benefits](https://scimeetings.acs.org/?utm_source=pubs_content_marketing&utm_medium=website&utm_campaign=0320_MCF_NPI_Launch_Spring_Homepage&ref=pubs_content_marketing)

[How to use SciMeetings](https://storage.googleapis.com/dl.morressier.com/acs/SciMeetings-ACSSpring2020-HowToGuide_M3020-Final.pdf)

[Frequently Asked Questions](https://storage.googleapis.com/dl.morressier.com/acs/SciMtng_ACSNMSp2020_FAQ_M3120_Final.pdf)

ACS Spring 2020 National Meeting & Expo

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.