Extended Data Fig. 8: The hexapod system facilitates precise and scalable identification of stimulatory CD4⁺ antigen-specific TCRs at the single-cell level.

a, Distribution of clonotype annotations within the repertoire, computed across the entire repertoire (left) and only among those with a productive TCR (right). When compared to the known 5C.C7 CDR3 amino acid (aa) sequence (alpha: AAEASNTNKVV; beta: ASSLNNANSDYT), each cell was classified as either a ‘5C.C7’ T cell (with the corresponding TCR alpha and/or TCR beta CDR3 aa sequence), an ‘other’ T cell (with a different productive CDR3 aa sequence), or ‘NA’ if a cell appears in the scRNA-seq data without the corresponding TCR information (potentially due to limited depth or dropout). b, Uniform Manifold Approximation and Projection (UMAP) showing single cells with the respective TCR clonotype annotations color-coded. c, Gene Set Enrichment Analyses (GSEA) showing selected pathways enriched in 5C.C7 cells compared to other cells. A positive enrichment score indicates enrichment in 5C.C7 cells over other cells, and a negative enrichment score implies enrichment in other cells compared to 5C.C7 cells. P-value was generated from a permutation test and adjusted for multiple testing using the Benjamini–Hochberg procedure. The false discovery rate (FDR) was estimated via the FDR q-values. d, Normalized gene expression of selected genes that are differentially expressed between 5C.C7 and other T cells. Group-wise mean expression was calculated as the average scaled gene expression of all cells within that group, and the final values were capped to maximize the dynamic range of the color scale. Genes were grouped into functional categories related to 5C.C7 cell TCR sequence or function.