Fig. 2: APMAT enables high-throughput representation of whole SARS-CoV-2 genome.

a Graphic relating the SARS-CoV-2 proteome to putative HLA A*02:01 restricted antigens (top row), to those SCT constructs that were expressed in usable yield (middle row), and those that captured antigen-specific CD8 T cells from COVID-19 participants (bottom row). For the SCT Expression row, darker red lines correspond to higher SCT expression, while gray means low/no expression. For the Cell Capture row, darker red lines mean more cells captured. b Bar plot showing the distribution of expressed SCTs for each SARS-COV-2 protein. SCTs that successfully captured CD8 T cells are shown in colored bars. The color code of the captured cells is that used in SARS-CoV-2 proteome of panel (a). c Antigen sequence motif of expressed (top left) vs non-expressed (bottom left) SCT constructs; and those expressed SCTs that did (top right) or did not capture CD8 T cells (bottom right). d For each peptide, conventional anchor positions (R2, R9) and non-anchor residues are assigned. The physicochemical properties are tabulated for each residue. Created in BioRender⁵⁹. e Bar plots comparing expressed (N = 560) vs non-expressed (N = 391) SCT constructs. X-axis: SCT expression status. Note that two of the expressed SCT constructs were not included for 10X experiment due to low sample volume. Y-axis: NetMHCpan prediction rank (left) and average hydrophobicity of exposed residues (right). f Bar plots comparing expressed SCTs that captured CD8 T cells (N = 102) vs those that did not (N = 456). X-axis: Cell capture status. Y-axis: NetMHCpan prediction rank (left) and average charge of exposed residues (right). Charge values represent the average of positive and negative charges rather than the absolute value. For bar plots, data are presented as mean values +/- SEM. The statistical significance was determined using the two-sided Mann-Whitney U test, and p-values are annotated on all relevant plots with exact p-values provided unless p < 0.0001.