Fig. 2

Properties of the nucleotide-level super-pangenome graph. a Compression ratio of the pangenome. For each step of the pangenome construction, raw (seqwish) and post-polishing (smoothxg), the mean ± sd (n = 19 chromosomes) of the total length of the unique nodes representing each chromosome is represented. The compression is expressed as a ratio of the total size of a chromosome in the graph compared with the sum of the lengths of the same chromosome in the 18 haplotypes used as input. b Graph-based pangenome modeling. For every combination of 1–9 genomes, the total length of the unique nodes is represented per class. The line represents smoothed conditional means with a 0.95 confidence interval. c Pangenome class composition per haplotype for each genome. d Repeat density observed in the different node classes of the pangenome. Significant groups were determined using a multiple comparison test after Kruskal–Wallis with a P value < 0.05. The middle bars represent the median while the bottom and top of each box represent the 25th and 75th percentiles, respectively. The whiskers extend to 1.5 times the interquartile range and data beyond the end of the whiskers are plotted individually as outlying points. Panels b, c, and d share the same color legend. e Transposable element composition in the different pangenome classes. Categories with unspecified repeats or representing < 2% of the repeat content were classified as “Other”