Extended Data Fig. 5: Structural interpretation of InDel fitness effects for 2A(pro) and 3D(pol).
(a) Structure of the EV-A71 2A(pro) (PDB: 3W95) with active site in yellow. (b) Cartoons showing the proteolytic activity of the 2A(pro) in processing the viral polyprotein and host proteins. (c) Area plot showing variants classified by Enrich2 score across the different secondary structure assignments in the 2A(pro). The width of the column represents the total number of residues that adopt a given secondary structure. The two-sided χ² statistics in panel (c) for 8 AA insertion χ² = 11.99 df = 4, 1 AA deletion χ² = 5.3477 df = 4, and AA change χ² = 63.698 df = 4; for 8 AA insertion: χ² = 73.753 df = 2, 1 AA deletion χ² = 2.6817 df = 2; and AA change: χ² = 32.706 df = 4. (d) Area plots showing the proportion of substitutions within Enrich2 classes for different secondary structures in the 2A(pro). (e) The structure of the EV-A71 3D(pol) with the GDD motif highlighted in yellow (PDB: 3N6L). (f) Cartoon showing the role of 3D(pol) in replication of viral genomes. (g) Area plot showing variants classified by Enrich2 score across the different secondary structure assignments in the 3D(pol). The width of the column represents the total number of residues that adopt a given secondary structure. The two-sided χ² statistics in panel (g) for 8 AA insertion χ² = 73.753 df = 2, 1 AA deletion χ² = 2.6817 df = 2, and AA change χ² = 32.706 df = 4. (h) Area plots showing the proportion of substitutions within Enrich2 classes for different secondary structures in the 3D(pol). (i) Heatmap showing the Enrich2 scores of the 3D(pol) contact residues with the template RNA. Black squares for AA substitutions represent the WT sequence. Black squares for deletions represent variants that were removed due to low variant counts in the input library.
