Fig. 5: Cell-type-specific ACRs are frequently species-specific.

a, A screenshot illustrating syntenic regions capturing shared ACRs across five species. The red bars denote syntenic ACRs within regions flanked by corresponding syntenic gene pairs, while the grey colour highlights these syntenic gene pairs. b, Three classes depicting variations in ACR conservation between two species. ‘Shared ACRs’, ACRs with matching sequences that are accessible in both species; ‘Variable ACRs’, ACRs with matching sequences but are only accessible in one species; ‘Species-specific ACRs’, ACRs where the sequence is exclusive to a single species. c, The percentage of broad and cell-type-specific ACRs underlying three classes shown in b. The significance test was done using one-tailed Fisher’s exact test (alternative = ‘greater’). d, Left, the number and percentage of O. sativa shared ACRs that retain or change cell-type specificity among the other four species. Right, a screenshot of an O. sativa phloem-specific ACR that retains phloem specificity in S. bicolor. This ACR is situated at the promoter region of LATERAL ROOT DEVELOPMENT 3 (LRD3) which is specifically expressed in companion cell and phloem sieve elements (Supplementary Table 2). The grey shaded region highlights the syntenic gene pair. e, The percentage of cell-type-specific ACRs identified across all cell types within species-specific class. The analysis demonstrates an enrichment of cell-type-specific ACRs based on Fisher’s exact test. This test assesses whether cell-type-specific ACRs are more likely to be situated in the ‘species-specific’ class, compared to other cell types. Statistically significant differences (P < 0.05 ranging from 0.023 to 0.005) in all pairwise comparisons are denoted by distinct letters, determined using one-tailed Fisher’s exact test with the alternative set to ‘greater’. Bars sharing the same letter indicate that they are not significantly (P > 0.05) different from each other. f, TF motif enrichment tests were performed in species-specific ACRs neighbouring O. sativa orthologue exhibiting higher expression levels in epidermis based on one-tailed binomial test (see ‘Binomial test-based motif enrichment analysis’ in the Methods). Significance testing in violin plot was performed using the t-test (alternative = ‘greater’). A total of 87 orthologous genes between Z. mays and O. sativa were used for the box plot analysis. g, GO enrichment test was performed in O. sativa orthologous genes based on agriGO¹⁴¹. h, A screenshot of LIP1 accessibility in O. sativa and Z. mays L1 cells which contains an O. sativa epidermal specific and species-specific ACR with two ZHD1 motif sites. No corresponding ZHD1 motifs were found in ZmLIP1.