Weekly Summary: Epigraph: A Vaccine Design Tool Applied to an HIV Therapeutic Vaccine and a Pan-Filovirus Vaccine, Sci Rep 2016.
The development of vaccines that are effective against diseases such as HIV and viruses belonging to the Filoviradae family as well as individually tailored therapeutic strategies for treatment of these diseases provides a difficult challenge. HIV especially presents challenges because of its high variability in the epitopes presented on its surface. Because of the uniqueness in epitope expression finding a single antigen vaccine for HIV has provide nearly impossible. Mosaic was a technique developed that incorporated a small set of antigens as a treatment that would improve coverage for these diverse epitopes by including a small subset, typically between 2 and 4, of epitopes based on a targeted population.
The authors hoped to improve upon Mosaic’s performance with the development of a new tool, Epigraph. Mosaic was designed using a genetic algorithm that was computationally expensive, to improve upon that epigraph was designed based on a graph system which greatly improved its performance in silico. The authors hypothesize that the epigraph method would provide therapeutic T-cell HIV vaccines that matched an individual’s infecting stain and aid in the development of a pan-filovirus vaccine that targeted all known viruses in the Filoviradae family.
To test this hypothesis, the authors applied their TTV solution to three potential HIV populations, a B-clade sequence sampled from the United States, a C-clade sequence sampled in Southern Africa and the 2015 Los Alamos HIV database global M-group Gag sequences. The authors then checked the results of this test for potential T-cell epitope coverage(PTE) as compared to non-tailored epigraph results and natural clade pairs. To test the second half of the hypothesis the authors used an aligned two-antigen Epigraph to identify conserved regions across all viruses within the Filoviradae family and based on these conserved regions and the additional constraints that the target vaccine preserved PTE coverage for EBOV, maintained excellent coverage for SUDV and MARV and finally while maintaining the previous two requirements still provided extensive PTE coverage for remaining species then ran a series of tests exploring potential design options. With these constraints in mind the authors then took a systematic approach to looking at PTE coverage based on a series of epigraph runs and demonstrated the potential to create a pan-Filoviradae vaccine that maintained high (>90%) PTE coverage against the most common of these viruses that still provided cross-reactive potential to very distinct alternative versions of the virus family.
Paper Critique: While this paper could demonstrate the potential of the algorithm and code for identification of unique epitopes and provide potential antigen combinations for vaccine development that has good PTE coverage, the lack of any live experiments to test these vaccines leave open the question of how useful this method as a practical means to design vaccines. Additionally, the paper could have benefited by providing more direct comparisons between the use of Epigraph and Mosaic for identification of PTEs.
Alternative Methods: An alternative computer technique utilizing agent modeling could potentially expand the concepts presented here. By defining the potential epitopes as agents and then allowing them to move around within a computational environment looking for different potential match partners; such as other agents (epitopes) or other environmental factors (defined as potentially other computers in a small cluster or other processes within a single computer) the potential exists to explore additional factors such as mutation and environment on epitope signatures.
Future direction: The performance enhancements Epigraph has shown over Mosaic have potential, but the results generated need to be tested in a similar manner as Mosaic results. A potential next step would involve the development of a pan-Filoviridae vaccine that is tested in an NHP population for its effectiveness and PTE coverage.