Fig 1.
Geographic, genetic and phenotypic diversity of 438 soybean accessions.
(A) Geographic and phenotypic distribution of 438 soybean cultivars. (B) Phylogenetic relationships among 438 soybean accessions. The seed color of each soybean accession is depicted in the second circle.
Fig 2.
GWAS analysis on the seed coat, hilum and pubescence color.
(A) Genome-wide Manhattan plots of the GWAS analyses. The statistical significance of P values is depicted by the negative logarithm with the threshold (gray dotted lines) of 5 x 10−9. Magnified Manhattan plots of candidate genomic region at the T locus (B), I locus (C) and R locus (D). Each colored circle denotes a different level of the SnpEff-defined (a simple putative impact assessment) functional impact of individual variants: red, high; orange, moderate; yellow, low; gray, modifier.
Table 1.
Information for key genes with the highest significance of variations at four seed color-associated loci.
Fig 3.
Genomic structures and gene expression profiles around the k-meri allele.
(A) Comparative genomic structures between G. soja (PI483463) and G. max (Williams 82 BAC77G7-a) at the I-cluster B. (B) Gene expression profile and Sashimi plot of chimeric subtilisin-CHS1/CHS3 gene. X-axis indicates relative genomic locations within the BAC77G7-a BAC clone, while Y-axis the read depth. Solid lines are used to depict exon/intron junction-to-junction, and corresponding number of reads that are splitted due to the junctions are denoted on each line. The upper part of Sashimi plot demonstrates mapping result of mRNA reads with an emphasis on distribution of junction-spanning reads along on the chimeric subtilisin-CHS1/CHS3. The dark and light colors in arrow-shaped boxes represent the exons and introns of the gene, respectively.
Fig 4.
Mapping analysis of small RNA-seq reads on the I-cluster B (A) and CHS7 (B). Read depth was counted by the number of reads on each position. Exon-intron junctions are denoted with the vertical dotted lines. The size and position of genes are all drawn to scale, while the read depth to the logarithmic scale.
Fig 5.
Schematic for the seed color-associated loci/genes within the context of the anthocyanin biosynthetic pathway.
Genes or corresponding enzymes are denoted with the capital letters of abbreviated names, as follows: CHS, chalcone synthase; F3’H, flavonoid 3’-hydroxylase; F3’5’H, flavonoid-3’,5’ hydroxylase; CHI, chalcone isomerase; DFR, dihydroflavonol-4-reductase; ANS, anthocyanidin synthase; UF3GT, flavonoid 3-O-glucosyltransferase; AOMT, anthocyanin O-methyltransferase.
Fig 6.
A proposed model for ‘Mirtron-Triggered Gene Silencing (MTGS)’ mechanism working on genome-wide CHS regulatory circuit.
It is noteworthy that the mirtron-derived primary miRNAs involve intron segments originated from CHS1/CHS3-containing transcript whereas the secondary siRNAs are all generated from the exon regions of other CHS transcripts. Abbreviations are as follows: RISC, RNA-induced silencing complex; AGO, argonaute proteins; RDR, RNA dependent RNA polymerase; DCL, dicer-like protein.