Fig. 1: Expression and methylation patterns in dry soybean seeds.

a Eleven soybean ecotypes were categorized into small-seed (depicted in green) and large-seed (depicted in red) groups based on seed dimensions, specifically length, width, and thickness. Scale bar represents 1 cm. b Measurements of length, width, thickness, and weight for 100 dry seeds were recorded. Data represent mean values with standard deviations (SDs) for at least 10 seeds from each ecotype (n = 10). Small-seed and large-seed groups are indicated by green and red points, respectively. c Gene Ontology (GO) enrichment analysis for downregulated genes between the two groups is shown, with molecular functions marked in blue and biological processes in yellow. The negative logarithm of the enrichment p-value visualizes the significance of differential expression. d A volcano plot illustrates the differentially expressed genes (DEGs) between the groups. Genes with a fold change exceeding 2 and an FDR below 0.01 are emphasized. Upregulated genes are marked with red dots, downregulated genes with blue, and GmDMEa and GmRDR2 with yellow. e Transcriptional variation of GmDMEa in dry soybean seeds. Transcript levels, quantified as FPKM, were derived from transcriptomic analyses of dry seeds. Data are presented in a box plot representation, where the interquartile range is delineated by the box, the median is indicated by the horizontal line within the box, and whiskers extend from the minimum to the maximum values. The small-seed cohort (n = 18; green box) includes ‘DN50’, ‘ZYD6’, ‘DN60’, ‘Dong Y’, ‘Bai YD’, and ‘Jack’, while the large-seed cohort (n = 15; red box) comprises ‘W82’, ‘SN14’, ‘HF51’, ‘DN42’, and ‘HF55’. Statistical significance was assessed by one-tailed, unpaired Student’s t-tests with a 95% confidence interval. Differences were considered statistically significant at p < 0.05. f Differential methylation patterns in dry soybean seeds. Methylation levels were quantified at CG, CHG, and CHH nucleotide contexts across gene bodies, intergenic regions, and transposable elements (TEs) within the soybean genome. The bar graph contrasts the methylation percentages between large (L) and small (S) seed groups. Differences were considered statistically significant at p < 0.05. g A stacked bar chart displays the counts of differentially methylated regions (DMRs) contrasting the large- and small-seed groups. Hypo-DMRs (lower methylation in the large-seed group) are shown in blue, and hyper-DMRs (higher methylation in the large-seed group) in orange. h A pie chart depicts the distribution of CHH-DMRs across the soybean genome, with Up_10000 and Down_10000 denoting regions 10,000 base pairs upstream and downstream from genes, respectively. i Average CHH methylation levels across four types of TEs in both small and large-seed groups are plotted in a line graph, with the small-seed group in green and the large-seed group in red.