Mutagenesis efficiency was assessed by scoring the occurrence of chlorotic and albino phenotypes. The observed rate of 0.6% of chlorotic and albino phenotypes in the 
mutant collection is in a similar range of previously described mutant collections and confirms the quality of the mutagenesis.. To validate the cucumber mutant collection, we screened for mutations in five genes and identified 26 independent alleles. As reported in other TILLinG studies, the EMS mutational specificity shows a strong preference for G/C to A/T transitions, 70 to 99% of the induced mutations. In our cucumber mutant population, most induced mutations were as expected, G/ C to A/T transitions, with the exception of the three following mutations, G/C to T/A, T/A to A/T and T/A to C/G. The spectrum of observed nucleotide changes is similar to the Oxytocin Syntocinon mutation spectrum observed in rice or tomato. Based on the TILLinG screens, we estimated the overall mutation density to one mutation every 1147 kb with an average of 5 alleles per gene. This mutation frequency is two fold lower than the rate reported for the closest cucurbit Cucumis melo, for tomatoand for sunflowerand equivalent to the rate of one mutation per megabase reported for barley. How the gender of a flower or a plant is determined is an important issue in plant developmental biology. Understanding this process also has practical applications in agriculture and plant breeding, as the gender of a flower or plant often limits how the plant is bred and cultivated. In cucumber, sex determination is genetically controlled by three master genes. We previously showed that the Monoeciouslocus in cucumber, is likely to encode for CsACS2. To test this hypothesis, we screened for induced mutations in CsACS2. Six independent mutations were identified and the mutant lines were backcrossed to the wild type and phenotyped. Detailed phenotypic characterization of the TILLinG mutants confirmed that Monoeciouslocus encodes for CsACS2. Interestingly, the only mutation, G33C, leading to sexual Riociguat BAY 63-2521 transition correspond to one of the natural mutations previously identified. This mutation is unlikely to be a contamination, as the G33C mutation was carried by the genetic background of Beit Alpha variety and the mutagenesis was carried out in controlled conditions. One explanation is that some genome sites are more susceptible to mutagenesis. Different studies reported biases of the EMS-induced mutation sites. The precise reason for the high mutability of specific sites is still unknown. However, we can speculate that highly exposed and not protected DNA sequences could be an easy target for guanine alkylation. Reduced DNA repair at certain sites could also leads to mutation hotspots. In conclusion, we have developed a reference EMS mutant collection and set up the cucumber TILLinG platform successfully. Through the TILLinG approach, we screened for induced mutations in the Monoecious sex determination gene, CsACS2 and showed that the G33C mutation leads to monoecy to andromonoecy sex transition. Cucumber is also an important model plant in many key areas of plant research, including fruit maturationand the investigation of vascular trafficking of molecules. Hence, by making the cucumber TILLinG platform available for the scientific community, we hope to fulfill the expectations of both breeders and scientists who are using cucumber as plant model.