Fig. 2: GWAS of P-acquisition efficiency in the natural population of soybean in the field.

a Histogram and density plot showing phenotypic distributions of plant P uptake. b Histogram and density plot showing phenotypic distributions of total root length. c Scatter plot showing phenotypic correlation of P uptake and total root length. Pearson correlation coefficient is shown and P value was calculated based on two-sided t-test. d Histogram and density plot showing phenotypic distributions of P-acquisition efficiency (P uptake per unit root length). Following a randomized complete block design in the field, averaged values of plant P content and root length of four biological replications for each accession at the seedling stage (one month after sowing) were calculated. P uptake per unit root length was calculated by dividing P content by total root length. e Manhattan plot (left) and quantile–quantile plot (right) of P-acquisition efficiency (P uptake per unit root length). n = 4 biologically independent experiments following a randomized complete block design in the field. For Manhattan plot, −log₁₀ P values are plotted against the position of SNPs on 20 chromosomes. Gray horizontal dashed line indicates genome-wide P value threshold determined by Bonferroni correction. The lead SNPs are shown above the corresponding association signals. For quantile–quantile plot, −log₁₀-transformed observed P values are plotted against -log₁₀-transformed expected P values. f Violin plots showing P-acquisition efficiency, P uptake, and yield of the two groups based on genotypes of the lead SNP sf2042902267. Points and error bars represent the averaged values and the standard deviation (SD) of each group, respectively. The P values were calculated using two-sided t-test. n = 49, 41; 52, 42; 40, 33 biologically independent samples, respectively. Source data underlying panels a–d, f are provided as a Source Data file.