Análisis comparativo de estrategias de genotipado masivo para la caracterización de poblaciones multi-parentales en trigo blando

Bread wheat is one of the most relevant crops on Earth, as it is the most cultivated crop with 220 million of hectares on 2020. The United Nations Organization stated that wheat accounts for 20% of the calories consumed by humans. Due to the increasing demand of wheat production and actual the climate change scenario, it is crucial any scientific research advance on wheat. Among the different wheat traits, quality, understood as both protein content and baking quality, is the most relevant trait to improve right behind yield. In order to carry out quality improvement in wheat varieties, it the crucial to identify the genes that control this complex trait, and the analysis of segregating populations is one of the most powerful tools for this purpose. Segregating populations can be divided in two groups, depending on the number of parentals used in its development: biparental and multiparent. This last group introduces a higher number of alleles as it uses multiple parents, allowing a greater amount of variability to be analyzed. The two most used types of multiparent populations are NAM (Nested Association Mapping populations) and MAGIC (Multi-parent Advanced Generation Intercross populations). In this project we will be focusing on NAM populations. In our case, multiple biparental populations where 11 Spanish bread wheat landraces will be crossed with a common parental, Chinese Spring. After developing the NAM populations, it is essential to have a good genetic characterization. During the last decade high-throughput technologies have changed plant breeding, the main genotyping systems can be classified in fixed arrays and NGS based technologies. To identify the most powerful genotyping technology, we analyzed the potential of a fixed array system, 90K Illumina ISelect that screens for around 90K SNP (Single Molecular Polymorphism) markers; and DArT-seq technology that uses Next Generation Sequencing and offers two types of datasets: PAV (presence or absence of variation) markers plus SNP. For the analysis the 12 parental lines were genotyped with both technologies and resulted data were processed using several scripts developed in R language. The identified polymorphic markers were compared between the three datasets DArT-PAVs, DArT-SNPs and 90K-SNPs by focusing on number of markers, their distribution among the chromosomes and the distance between them. A total of 131042 markers for DArT-PAV, 58660 for DArT-SNP and 8155 5 for 90K Array were obtained. After selecting the polymorphic between each of the Spanish bread wheat landraces and Chinese spring, an average of 22666, 6294 and 3032 markers remained respectively. When we analyzed the density of markers and distances between them the potential superiority of the DArT-seq technology was clear. Actually, the 90K Illumina array showed lower number of polymorphic markers and huge genetic distances between them markers. Moreover, the combination of both DArT-seq markers types (PAVs plus SNPs) resulted in an increase of number of polymorphic markers and a reduction of the distances between them, providing it to be the best strategy for genotyping the future NAM populations. After determining the potential of the DArT-seq technology, a set of DArT-seq SNP markers were selected for monitoring the NAM population development. These 7 new markers and 9 preexisting markers (previously developed by our research group) were set up with DNA samples from the 12 parental lines and 10 of the 11 F1 hybrids obtained from crossing them. Results showed that only 2 of the 16 markers were inaccurate, and that the other 14 were enough to precisely identified the different p<rental lines and the F1s. These results will allow to supervise the development of the NAM populations, that when completed will be high-throughput genotyped by DArT-seq in order to construct genetic maps and to identify the genes controlling the different traits segregating.