Fig. 2: Large Scale of Streptactin-binding Peptides from Deep RD.

a Protein structure in RD screening. Myc tag equalizes translation levels; 12×NNK random peptides; Gene3 spacer; SecM sequence enables ribosome stalling and stable ternary complex formation. b Statistics at each step of the standard sequence extraction pipeline. 243.25 million NGS reads containing 73.42GBase were processed to extract the inserted NNK region, obtaining 221.99 million high-quality reads for further analysis. c Sequencing data statistics after five rounds of RD screening. Total sequencing reads and unique amino acid sequence variants after each screening round. While read counts remained similar across rounds, type number of variants decreased significantly with screening progression, indicating sequence enrichment. d Venn diagram illustrating the distribution of sequence sets across screening stages: the gray and brown circles represent the initial library and the sequences after the first round of screening, respectively; the red area denotes the union of sequences from the second to fifth rounds of screening. The right subpanel shows an independent Venn diagram of the results from rounds 2, 3, 4, and 5, with overlapping regions representing shared sequences between rounds. e Distribution of sequencing depth after five rounds of RD screening. Sequence depth (x-axis) versus cumulative frequency (y-axis). The proportion of high-depth sequences increases with screening rounds (0~5, color gradient from blue to red). Both axes are scaled logarithmically. f Correlation analysis of sequence depth across five rounds of RD screening. Heatmap of Pearson’s correlation coefficients between sequencing depths in rounds 1~5. Round 1 showed low correlation with other rounds (r≈0.00), while rounds 2~5 exhibited progressively stronger correlations (r = 0.48~0.91). Color gradient indicates correlation strength (blue: low; red: high).