Upland cotton is a model for polyploid crop domestication and transgenic improvement. Here we sequenced the allotetraploid Gossypium hirsutum L. acc. TM-1 genome by integrating whole-genome shotgun reads, bacterial artificial chromosome (BAC)-end sequences and genotype-by-sequencing genetic maps. We assembled and annotated 32,032 A-subgenome genes and 34,402 D-subgenome genes. Structural rearrangements, gene loss, disrupted genes and sequence divergence were more common in the A subgenome than in the D subgenome, suggesting asymmetric evolution. However, no genome-wide expression dominance was found between the subgenomes. Genomic signatures of selection and domestication are associated with positively selected genes (PSGs) for fiber improvement in the A subgenome and for stress tolerance in the D subgenome. This draft genome sequence provides a resource for engineering superior cotton lines.

High-throughput genotyping arrays provide a standardized resource for plant breeding communities that are useful for a breadth of applications including high-density genetic mapping, genome-wide association studies (GWAS), genomic selection (GS), complex trait dissection, and studying patterns of genomic diversity among cultivars and wild accessions. We have developed the CottonSNP63K, an Illumina Infinium array containing assays for 45,104 putative intraspecific single nucleotide polymorphism (SNP) markers for use within the cultivated cotton species Gossypium hirsutum L. and 17,954 putative interspecific SNP markers for use with crosses of other cotton species with G. hirsutum. The SNPs on the array were developed from 13 different discovery sets that represent a diverse range of G. hirsutum germplasm and five other species: G. barbadense L., G. tomentosum Nuttal × Seemann, G. mustelinum Miers × Watt, G. armourianum Kearny, and G. longicalyx J.B. Hutchinson and Lee. The array was validated with 1,156 samples to generate cluster positions to facilitate automated analysis of 38,822 polymorphic markers. Two high-density genetic maps containing a total of 22,829 SNPs were generated for two F₂ mapping populations, one intraspecific and one interspecific, and 3,533 SNP markers were co-occurring in both maps. The produced intraspecific genetic map is the first saturated map that associates into 26 linkage groups corresponding to the number of cotton chromosomes for a cross between two G. hirsutum lines. The linkage maps were shown to have high levels of collinearity to the JGI G. raimondii Ulbrich reference genome sequence. The CottonSNP63K array, cluster file and associated marker sequences constitute a major new resource for the global cotton research community.

Linked-Read sequencing technology has recently been employed successfully for de novo assembly of human genomes, however, the utility of this technology for complex plant genomes is unproven. We evaluated the technology for this purpose by sequencing the 3.5-gigabase (Gb) diploid pepper (Capsicum annuum) genome with a single Linked-Read library. Plant genomes, including pepper, are characterized by long, highly similar repetitive sequences. Accordingly, significant effort is used to ensure that the sequenced plant is highly homozygous and the resulting assembly is a haploid consensus. With a phased assembly approach, we targeted a heterozygous F₁ derived from a wide cross to assess the ability to derive both haplotypes and characterize a pungency gene with a large insertion/deletion. The Supernova software generated a highly ordered, more contiguous sequence assembly than all currently available C. annuum reference genomes. Over 83% of the final assembly was anchored and oriented using four publicly available de novo linkage maps. A comparison of the annotation of conserved eukaryotic genes indicated the completeness of assembly. The validity of the phased assembly is further demonstrated with the complete recovery of both 2.5-Kb insertion/deletion haplotypes of the PUN1 locus in the F₁ sample that represents pungent and nonpungent peppers, as well as nearly full recovery of the BUSCO2 gene set within each of the two haplotypes. The most contiguous pepper genome assembly to date has been generated which demonstrates that Linked-Read library technology provides a tool to de novo assemble complex highly repetitive heterozygous plant genomes. This technology can provide an opportunity to cost-effectively develop high-quality genome assemblies for other complex plants and compare structural and gene differences through accurate haplotype reconstruction.

Expression quantitative trait loci (eQTL) studies have established convincing relationships between genetic variants and gene expression. Most of these studies focused on the mean of gene expression level, but not the variance of gene expression level (i.e., gene expression variability). In the present study, we systematically explore genome-wide association between genetic variants and gene expression variability in humans. We adapt the double generalized linear model (dglm) to simultaneously fit the means and the variances of gene expression among the three possible genotypes of a biallelic SNP. The genomic loci showing significant association between the variances of gene expression and the genotypes are termed expression variability QTL (evQTL). Using a data set of gene expression in lymphoblastoid cell lines (LCLs) derived from 210 HapMap individuals, we identify cis-acting evQTL involving 218 distinct genes, among which 8 genes, ADCY1, CTNNA2, DAAM2, FERMT2, IL6, PLOD2, SNX7, and TNFRSF11B, are cross-validated using an extra expression data set of the same LCLs. We also identify ∼300 trans-acting evQTL between >13,000 common SNPs and 500 randomly selected representative genes. We employ two distinct scenarios, emphasizing single-SNP and multiple-SNP effects on expression variability, to explain the formation of evQTL. We argue that detecting evQTL may represent a novel method for effectively screening for genetic interactions, especially when the multiple-SNP influence on expression variability is implied. The implication of our results for revealing genetic mechanisms of gene expression variability is discussed.

Understanding the composition, evolution, and function of the Gossypium hirsutum (cotton) genome is complicated by the joint presence of two genomes in its nucleus (AT and DT genomes). These two genomes were derived from progenitor A-genome and D-genome diploids involved in ancestral allopolyploidization. To better understand the allopolyploid genome, we re-sequenced the genomes of extant diploid relatives that contain the A1 (Gossypium herbaceum), A2 (Gossypium arboreum), or D5 (Gossypium raimondii) genomes. We conducted a comparative analysis using deep re-sequencing of multiple accessions of each diploid species and identified 24 million SNPs between the A-diploid and D-diploid genomes. These analyses facilitated the construction of a robust index of conserved SNPs between the A-genomes and D-genomes at all detected polymorphic loci. This index is widely applicable for read mapping efforts of other diploid and allopolyploid Gossypium accessions. Further analysis also revealed locations of putative duplications and deletions in the A-genome relative to the D-genome reference sequence. The approximately 25,400 deleted regions included more than 50% deletion of 978 genes, including many involved with starch synthesis. In the polyploid genome, we also detected 1,472 conversion events between homoeologous chromosomes, including events that overlapped 113 genes. Continued characterization of the Gossypium genomes will further enhance our ability to manipulate fiber and agronomic production of cotton.

Cotton germplasm resources contain beneficial alleles that can be exploited to develop germplasm adapted to emerging environmental and climate conditions. Accessions and lines have traditionally been characterized based on phenotypes, but phenotypic profiles are limited by the cost, time, and space required to make visual observations and measurements. With advances in molecular genetic methods, genotypic profiles are increasingly able to identify differences among accessions due to the larger number of genetic markers that can be measured. A combination of both methods would greatly enhance our ability to characterize germplasm resources. Recent efforts have culminated in the identification of sufficient SNP markers to establish high-throughput genotyping systems, such as the CottonSNP63K array, which enables a researcher to efficiently analyze large numbers of SNP markers and obtain highly repeatable results. In the current investigation, we have utilized the SNP array for analyzing genetic diversity primarily among cotton cultivars, making comparisons to SSR-based phylogenetic analyses, and identifying loci associated with seed nutritional traits. The SNP markers distinctly separated G. hirsutum from other Gossypium species and distinguished the wild from cultivated types of G. hirsutum. The markers also efficiently discerned differences among cultivars, which was the primary goal when designing the CottonSNP63K array. Population structure within the genus compared favorably with previous results obtained using SSR markers, and an association study identified loci linked to factors that affect cottonseed protein content. Our results provide a large genome-wide variation data set for primarily cultivated cotton. Thousands of SNPs in representative cotton genotypes provide an opportunity to finely discriminate among cultivated cotton from around the world. The SNPs will be relevant as dense markers of genome variation for association mapping approaches aimed at correlating molecular polymorphisms with variation in phenotypic traits, as well as for molecular breeding approaches in cotton.