Algorithmic Bioinformatics, Heinrich Heine University Düsseldorf, Düsseldorf, Germany.
Institute of Medical Microbiology and Hospital Hygiene, University Clinic Düsseldorf, Düsseldorf, Germany.
Cell Syst. 2023 Dec 20;14(12):1122-1130.e3. doi: 10.1016/j.cels.2023.11.001.
The efficacy of epitope vaccines depends on the included epitopes as well as the probability that the selected epitopes are presented by the major histocompatibility complex (MHC) proteins of a vaccinated individual. Designing vaccines that effectively immunize a high proportion of the population is challenging because of high MHC polymorphism, diverging MHC-peptide binding affinities, and physical constraints on epitope vaccine constructs. Here, we present HOGVAX, a combinatorial optimization approach for epitope vaccine design. To optimize population coverage within the constraint of limited vaccine construct space, HOGVAX employs a hierarchical overlap graph (HOG) to identify and exploit overlaps between selected peptides and explicitly models the structure of linkage disequilibrium in the MHC. In a SARS-CoV-2 case study, we demonstrate that HOGVAX-designed vaccines contain substantially more epitopes than vaccines built from concatenated peptides and predict vaccine efficacy in over 98% of the population with high numbers of presented peptides in vaccinated individuals.
表位疫苗的功效取决于所包含的表位以及所选表位被接种个体的主要组织相容性复合体 (MHC) 蛋白呈现的概率。由于 MHC 多态性高、MHC-肽结合亲和力不同以及表位疫苗构建的物理限制，设计能有效免疫很大一部分人群的疫苗具有挑战性。在这里，我们提出了 HOGVAX，这是一种用于表位疫苗设计的组合优化方法。为了在有限的疫苗构建空间的限制内优化人群覆盖范围，HOGVAX 使用层次重叠图 (HOG) 来识别和利用所选肽之间的重叠，并明确模拟 MHC 中连锁不平衡的结构。在 SARS-CoV-2 的案例研究中，我们证明了 HOGVAX 设计的疫苗包含的表位数量大大超过了由串联肽构建的疫苗，并且预测了疫苗在接种个体中呈现的肽数量高的情况下超过 98%的人群中的功效。
