Utilizing Computational Machine Learning Tools to Understand Immunogenic Breadth in the Context of a CD8 T-Cell Mediated HIV Response

Abstract

Predictive models are becoming more and more commonplace as tools for candidate antigen discovery to meet the challenges of enabling epitope mapping of cohorts with diverse HLA properties. Here we build on the concept of using two key parameters, diversity metric of the HLA profile of individuals within a population and consideration of sequence diversity in the context of an individuals CD8 T-cell immune repertoire to assess the HIV proteome for defined regions of immunogenicity. Using this approach, Analysis of HLA adaptation and functional immunogenicity data enabled the identification of regions within the proteome that offer significant conservation, HLA recognition within a population, low prevalence of HLA adaptation and demonstrated immunogenicity. We believe this unique and novel approach to vaccine design that, in combination with in vitro functional assays, offers a bespoke pipeline for expedited and rational CD8 T-cell vaccine design for HIV and potentially other pathogens with the potential for both global and local coverage.

Direct identification of HLA‐presented CD8 T cell epitopes from transmitted founder HIV‐1 variants

“…HIV variants with different phenotypic characteristics (replicative capacity, sequence diversity, clade) can be used to infect PBMC samples derived from HIV naïve individuals. The distribution of HLA allele frequency has been assessed previously including the prevalence of different HLA alleles within a longitudinal African HIV cohort [22].…”

Selective HLA restriction enables the evaluation and interpretation of immunogenic breadth at comparable levels to that observed with broader HLA distribution

“…As a theoretical proof of concept for this technique to characterize HLA diversity, a subset of 13 volunteers enrolled within IAVI Protocol C, a longitudinal prospective study of 613 HIV positive volunteers [11,12], were evaluated for their representative frequency and distribution against the full cohort by 4 digit characterization and HLA binding profile. The data indicated that at the level of allele frequency, these volunteers were representing >80% of total HLA-A, -B and -C frequency [13]. An HLA binding profile was then computed for each allele by predicting the binding affinity for 9mer HIV gag peptides against the binding profile of the total HLA allele distribution from within Protocol C. From these profiles, the overall binding characteristics of each volunteer based on their HLA restriction were compiled.…”

“…By evaluating the different affinities calculated by the algorithm, it is possible to assign a distance between each allele and each volunteer based on their predicted ability to iteratively bind gag peptides. The distances are then used to cluster alleles/volunteers and visualize HLA diversity [13]. Utilizing these concepts, it is then possible to identify, within a population, individuals with different allele frequencies and distributions that can represent the total HLA diversity within the same population.…”

Sampling SARS-CoV-2 Proteomes for Predicted CD8 T-Cell Epitopes as a Tool for Understanding Immunogenic Breadth and Rational Vaccine Design

“…Previously we have applied NetMHCpan ( Nielsen and Andreatta, 2016 ) as a proxy to identify putative CD8 T-cell epitopes contained within the HIV transmitted founder virus (TFV) identified from the Protocol C clinical cohort of sub Saharan and East Africa. We have shown that it is possible to stratify and rank protein and/or proteome sequences for their contributions of potential T-cell epitopes ( McGowan et al, 2020 ). Here we propose to use the same approach to evaluate a subset of global circulating SARS-CoV-2 sequences and, using the predefined analysis applied to modeling HIV diversity, identify key regions within the SARS-CoV-2 proteome that could be included within a therapeutic T-cell vaccine.…”