Domestication modifies the genomic variation of species. Quantifying this variation provides insights into the domestication process, facilitates the management of resources used by breeders and germplasm centers, and enables the design of experiments to associate traits with genes. We described and analyzed the genetic diversity of 1,008 tomato accessions including Solanum lycopersicum var. lycopersicum (SLL), S. lycopersicum var. cerasiforme (SLC), and S. pimpinellifolium (SP) that were genotyped using 7,720 SNPs. Additionally, we explored the allelic frequency of six loci affecting fruit weight and shape to infer patterns of selection. Our results revealed a pattern of variation that strongly supported a two-step domestication process, occasional hybridization in the wild, and differentiation through human selection. These interpretations were consistent with the observed allele frequencies for the six loci affecting fruit weight and shape. Fruit weight was strongly selected in SLC in the Andean region of Ecuador and Northern Peru prior to the domestication of tomato in Mesoamerica. Alleles affecting fruit shape were differentially selected among SLL genetic subgroups. Our results also clarified the biological status of SLC. True SLC was phylogenetically positioned between SP and SLL and its fruit morphology was diverse. SLC and “cherry tomato” are not synonymous terms. The morphologically-based term “cherry tomato” included some SLC, contemporary varieties, as well as many admixtures between SP and SLL. Contemporary SLL showed a moderate increase in nucleotide diversity, when compared with vintage groups. This study presents a broad and detailed representation of the genomic variation in tomato. Tomato domestication seems to have followed a two step-process; a first domestication in South America and a second step in Mesoamerica. The distribution of fruit weight and shape alleles supports that domestication of SLC occurred in the Andean region. Our results also clarify the biological status of SLC as true phylogenetic group within tomato. We detect Ecuadorian and Peruvian accessions that may represent a pool of unexplored variation that could be of interest for crop improvement.

Summary: The Cucurbita genus (squashes, pumpkins and gourds) includes important domesticated species such as C. pepo, C. maxima and C. moschata. In this study, we present a high‐quality draft of the zucchini (C. pepo) genome. The assembly has a size of 263 Mb, a scaffold N50 of 1.8 Mb and 34 240 gene models. It includes 92% of the conserved BUSCO core gene set, and it is estimated to cover 93.0% of the genome. The genome is organized in 20 pseudomolecules that represent 81.4% of the assembly, and it is integrated with a genetic map of 7718 SNPs. Despite the small genome size, three independent lines of evidence support that the C. pepo genome is the result of a whole‐genome duplication: the topology of the gene family phylogenies, the karyotype organization and the distribution of 4DTv distances. Additionally, 40 transcriptomes of 12 species of the genus were assembled and analysed together with all the other published genomes of the Cucurbitaceae family. The duplication was detected in all the Cucurbita species analysed, including C. maxima and C. moschata, but not in the more distant cucurbits belonging to the Cucumis and Citrullus genera, and it is likely to have occurred 30 ± 4 Mya in the ancestral species that gave rise to the genus.

Cucurbita pepo is a cucurbit with growing economic importance worldwide. Zucchini morphotype is the most important within this highly variable species. Recently, transcriptome and Simple Sequence Repeat (SSR)- and Single Nucleotide Polymorphism (SNP)-based medium density maps have been reported, however further genomic tools are needed for efficient molecular breeding in the species. Our objective is to combine currently available complete transcriptomes and the Zucchini genome sequence with high throughput genotyping methods, mapping population development and extensive phenotyping to facilitate the advance of genomic research in this species. We report the Genotyping-by-sequencing analysis of a RIL population developed from the inter subspecific cross Zucchini x Scallop (ssp. pepo x ssp. ovifera). Several thousands of SNP markers were identified and genotyped, followed by the construction of a high-density linkage map based on 7,718 SNPs (average of 386 markers/linkage group) covering 2,817.6 cM of the whole genome, which is a great improvement with respect to previous maps. A QTL analysis was performed using phenotypic data obtained from the RIL population from three environments. In total, 48 consistent QTLs for vine, flowering and fruit quality traits were detected on the basis of a multiple-environment analysis, distributed in 33 independent positions in 15 LGs, and each QTL explained 1.5–62.9% of the phenotypic variance. Eight major QTLs, which could explain greater than 20% of the phenotypic variation were detected and the underlying candidate genes identified. Here we report the first SNP saturated map in the species, anchored to the physical map. Additionally, several consistent QTLs related to early flowering, fruit shape and length, and rind and flesh color are reported as well as candidate genes for them. This information will enhance molecular breeding in C. pepo and will assist the gene cloning underlying the studied QTLs, helping to reveal the genetic basis of the studied processes in squash.

A collection of 163 accessions, including Solanum pimpinellifolium, Solanum lycopersicum var. cerasiforme and Solanum lycopersicum var. lycopersicum, was selected to represent the genetic and morphological variability of tomato at its centers of origin and domestication: Andean regions of Peru and Ecuador and Mesoamerica. The collection is enriched with S. lycopersicum var. cerasiforme from the Amazonian region that has not been analyzed previously nor used extensively. The collection has been morphologically characterized showing diversity for fruit, flower and vegetative traits. Their genomes were sequenced in the Varitome project and are publicly available (solgenomics.net/projects/varitome). The identified SNPs have been annotated with respect to their impact and a total number of 37,974 out of 19,364,146 SNPs have been described as high impact by the SnpEeff analysis. GWAS has shown associations for different traits, demonstrating the potential of this collection for this kind of analysis. We have not only identified known QTLs and genes, but also new regions associated with traits such as fruit color, number of flowers per inflorescence or inflorescence architecture. To speed up and facilitate the use of this information, F2 populations were constructed by crossing the whole collection with three different parents. This F2 collection is useful for testing SNPs identified by GWAs, selection sweeps or any other candidate gene. All data is available on Solanaceae Genomics Network and the accession and F2 seeds are freely available at COMAV and at TGRC genebanks. All these resources together make this collection a good candidate for genetic studies.