Extended Data Fig. 7: Identification of candidate genes involved in improvement breeding.

a-c, LL1p6. d-f, LL9q21. g-h, GNP3p1. i-l, GT8p4. m-t, GW5. (a, d, g, i, k, m, o, q, s) Local manhattan plots surrounding the genomic regions where the candidate loci are located. (b, e, h, j, l, n, p, r, t) boxplots or barplot demonstrating the phenotypic observation distribution of 3 genotypes. Peak association signals marked in blue dots were used for genotyping. n = sample size. Significant test was conducted by one-way ANOVA for data with homoscedasticity distribution or Kruskal-Wallis test for data with heteroscedasticity distribution. And multiple comparison was further conducted by the least significant difference (LSD) method with ‘Bonferroni’ correction for homoscedasticity distribution or Nemenyi test for heteroscedasticity distribution. Different lowercase letters above the boxplots indicate significant phenotype difference (p ≤ 0.05). (c and f) Two newly-identified loci in GWAS were also identified by the linkage mapping of leaf length in two sets of segregating populations, respectively. LL1p6 was identified in a F₂ population from the indica-indica hybrid Longjingyou534 and LL9q21 was from the indica-indica hybrid Quanliangyou2118. LOD (likelihood of odds) values were plotted against the physical positions. The threshold was set as 2.5, and was indicated by a dashed line.