Fig. 5

Human nullpeptide characterization. a Nullpeptide prioritization based on the number of occurrences of all permuted peptides of length 4aa (top) and 5aa (bottom). Black line indicates the mean number of occurrences of all permuted peptides with upper and lower lines indicating minimum and maximum number of occurrences across the permuted peptides. Nullpeptides were ordered based on the average number of times they were found in the simulations. b φ2 metric score number of occurrences of nullpeptides in the proteome relative to their occurrences in simulations controlled for mono-, di-, and tripeptide content of the proteome plotted as a function of nullomer length for K = 10–15 bp. Purple, turquoise, and red colors represent occurrences of nullpeptides in the simulations controlling mono-, di-, and tripeptide content, respectively. c Simulations showing that a large proportion of peptide nullomers (K = 4aa and K = 5aa) should be frequently observed in the human proteome and are thus likely under negative selection. d Depiction of the twenty-five most frequent nullomers in the simulations for K = 4 and K = 5 aa. e, f Number of occurrences of permuted peptides for each nullpeptide for K = 4 aa and K = 5aa. Black line represents the median occurrences across the permuted peptides. Only the twenty-five top ranked nullpeptides of 4- and 5-amino acid length are shown. g Number of species in which human nullpeptides of 4–5-amino acid length were identified. h Distribution of nullpeptides across the number of all possible substitutions that allow a nullpeptide to materialize. i Gene Ontology (GO) analysis of the top and bottom 10% genes based on the putative nullpeptide-resurfacing mutation density for K = 5aa nullpeptides. The circle diameter correlates to the number of contributing genes and the circle shade correlates to GO term q-value