ABSTRACT
Melon (Cucumis melo) and summer squash (Cucurbita pepo) are economically important cucurbits cultivated worldwide. New, better-adapted, cultivars, resistant to pests and diseases, and varieties displaying better quality traits are required to optimize their production and to meet market demands. Crop breeding must be carried out efficiently and competitively, using the existing genetic knowledge and the molecular tools available for each species. The development of genomic tools to speed up Cucurbits breeding programs is the main objective of this Doctoral Thesis.
Molecular markers are essential tools for the construction of genetic maps, for an efficient MAS selection, for QTL (Quantitative Trait Loci) analysis and mapping and for the development of new breeding populations. Molecular markers are also necessary tools for biodiversity analysis. High-quality markers, such as microsatellites (Simple Sequence Repeats, SSRs) and SNPs (Single Nucleotide Polymorphisms) have been developed and/or validated for both species in this Thesis. The generation of sequence information, required for the development of SSRs and SNPs has dramatically changed in the last 6 years due to the advent of new technological advances. In fact, a fairly complete melon transcriptome has been produced using high-throughput 454 sequencing (NGS, Next Generation Sequencing) and combining the newly produced ESTs with the previously existing Sanger sequences. The melon transcriptome characterization presented is the most complete to date (53.252 unigenes with over 63% of them annotated). The generation of large SSRs and SNPs collections for these species, by mapping the ESTs (Expressed Sequence Tags) against the new version of the transcriptome has been a great achievement starting a new era in melon breeding. The obtained collection of molecular markers (more than 3.000 SSRs and 38.000 SNPs) has allowed the construction of melon maps with increased marker density.
Similar resources have been developed for the first time in C. pepo. SSRs identified in the C. pepo transcriptome sequencing project has been applied to localize landraces with favourable commercial traits and whose alleles might be exploited in breeding programs. Likewise, the usefulness of both EST-SSRs and genomic SSRs for depicting genetic relationships has been proved. On the other hand, the employment of the first SNP set identified for Cucurbita through a massive genotyping platform, GoldenGate, has been a success (90%). The genetic map obtained, with more than 300 SNPs, has allowed to map a large set of QTLs related to early flowering, femaleness tendency, shape, size and fruit color, and other traits of interest in breeding. 
Therefore, these works report the validation and application of previously developed markers and the generation of new genomic tools to speed up the breeding process of these crops. The paper-based format of this Thesis confirms the interest of the obtained results.