An Optimal Strategy for Using Mixture-Based Combinatorial Library Data to Find T-Cell Epitopes
Abstract
Positional scanning libraries are well established in their use for peptide sequence identification and binding site optimization. Their use in identifying T cell stimulatory peptides has also previously been explored. Here we present a retrospective analysis on the results of mixture based positional scanning libraries of T-cell stimulatory peptides done before 2019 in order to identify an optimal strategy for using mixture based combinatorial library data to find T cell epitopes. The screening of positional scanning libraries against CD4+ and CD8+ T cell clones was analyzed using biometrical analysis approaches of various kinds in order to evaluate their relative effectiveness for finding antigens of unknown specificity. We found that, in general, these multiple different methods resulted in strikingly similar success rates. Notable results include the top 25 logsum scores of CD8+ T cells have a success rate which was almost double that of the top 25 sum scores. Logsum analysis of the top 25 CD8+ T cells was identical to that of the top 50 and only slightly higher for the top 100. By determining the optimal approach, scientists can better utilize mixture-based combinatorial library data to identify optimal peptide sequences.
Faculty Sponsors
Dr. Radleigh Santos
Project Type
Event
Location
Alvin Sherman Library
Start Date
4-5-2023 12:00 PM
End Date
4-6-2023 4:00 PM
An Optimal Strategy for Using Mixture-Based Combinatorial Library Data to Find T-Cell Epitopes
Alvin Sherman Library
Positional scanning libraries are well established in their use for peptide sequence identification and binding site optimization. Their use in identifying T cell stimulatory peptides has also previously been explored. Here we present a retrospective analysis on the results of mixture based positional scanning libraries of T-cell stimulatory peptides done before 2019 in order to identify an optimal strategy for using mixture based combinatorial library data to find T cell epitopes. The screening of positional scanning libraries against CD4+ and CD8+ T cell clones was analyzed using biometrical analysis approaches of various kinds in order to evaluate their relative effectiveness for finding antigens of unknown specificity. We found that, in general, these multiple different methods resulted in strikingly similar success rates. Notable results include the top 25 logsum scores of CD8+ T cells have a success rate which was almost double that of the top 25 sum scores. Logsum analysis of the top 25 CD8+ T cells was identical to that of the top 50 and only slightly higher for the top 100. By determining the optimal approach, scientists can better utilize mixture-based combinatorial library data to identify optimal peptide sequences.
