Genomic Insight into Differentiation and Selection Sweeps in the Improvement of Upland Cotton

Abstract

Upland cotton is the most economically important fibre crop. The human-mediated selection has resulted in modern upland cultivars with higher yield and better fibre quality. However, changes in genome structure resulted from human-mediated selection are poorly understood. Comparative population genomics offers us tools to dissect the genetic history of domestication and helps to understand the genome-wide effects of human-mediated selection. Hereby, we report a comprehensive assessment of Gossypium hirsutum landraces, obsolete cultivars and modern cultivars based on high throughput genome-wide sequencing of the core set of genotypes. As a result of the genome-wide scan, we identified 93 differential regions and 311 selection sweeps associated with domestication and improvement. Furthermore, we performed genome-wide association studies to identify traits associated with the differential regions and selection sweeps. Our study provides a genetic basis to understand the domestication process in Chinese cotton cultivars. It also provides a comprehensive insight into changes in genome structure due to selection and improvement during the last century. We also identified multiple genome-wide associations (GWAS associations) for fibre yield, quality and other morphological characteristics.

Parallel and Intertwining Threads of Domestication in Allopolyploid Cotton

“…genomes [41] and echoes earlier results. [27] Notably, the number of SNPs identified here is three-to-15-fold greater than found in previous studies, [24][25][26][27]32,34] due to the greater sequencing depth and breadth of sampling. S1, and Figure S2, Supporting Information).…”

“…A distinguishing feature of our study is that the sampling strategy focused on the inclusion of wild forms or variously primitive landrace accessions, thus enabling insight into the patterns and processes accompanying crop domestication and improvement. These data were combined with existing data for hundreds of other accessions used in previous studies [24][25][26][27]32,34] to generate a data set that, following quality filtering, yielded 23.0 M and 26.6 M high-quality SNPs in G. hirsutum and G. barbadense, respectively, with these two species being distinguished by 33.8M SNPs. As detailed below, these data provide insights into A) phylogenetic relationships among and within species; B) patterns of genetic diversity and the impact of domestication; C) the amount and genomic distribution of reciprocal interspecific introgression between the two cultivated species; and D) signatures of selection accompanying crop improvement.…”

“…These results parallel those of Ma et al [27] who reported a similar asymmetry for fiber-related genes, but differ from other results using different germplasm comparisons. [32] Notably, expression levels for 157 genes in these putatively selected regions were altered by domestication (Note S4, Table S3, and Figure S3, Supporting Information), thereby providing candidates for further functional relationships to the radically altered morphology of wild versus domesticated cotton fiber. In contrast to G. hirsutum, regions of selection detected in the A T -genome of G. barbadense are twice in number and total length relative to D T (168 and 48.34 MB in A T vs 93 and 21.54 MB in D T , respectively Note S4 and Table S5, Supporting Information).…”

“…[11][12][13][14] Whole-genome resequencing has been used to decipher the underpinnings of the domestication syndrome in many crops, including rice, [15][16][17] maize, [18] soybean, [19] and others. Recent advances in cotton genome sequencing [20][21][22][23][24][25][26][27][28][29][30][31][32][33] have been used to assess the domestication history of G. hirsutum and G. barbadense; however, these studies included limited representation of the wild/landrace gene pool from the ancestral geographic centers of origin and/or low coverage sequencing. Accordingly, the present study was designed expressly to assess the effects of domestication and improvement of G. hirsutum and G. barbadense on the origin and composition of their modern gene pools.…”

Genome-wide association analysis reveals quantitative trait loci and candidate genes involved in yield components under multiple field environments in cotton (Gossypium hirsutum)

“…Therefore, to improve the authenticity and validity of QTLs, we selected QTLs detected in two or more environments as candidate QTLs. In the 95 candidate QTLs, 38 co-located QTLs have been identified in multiple reported studies [ 12 , 13 , 15 , 20 , 21 , 23 , 29 , 30 ]. In addition, 57 novel QTLs were detected in multiple environments; for example, qtl83 was associated with BW, LP and SI in nine environments, providing more loci and markers for genetic improvement of cotton yield traits.…”