Identificação de QTLs em soja associados à resistência aos percevejos e a caracteres agronômicos utilizando a abordagem de mapeamento multivariado

Abstract

QTL identification in soybean related to stink bug resistance and agronomic traits using the multivariate multiple interval mapping approach Soybean is the Brazilian crop with the most expansion along the past three decades. Currently, it is a commodity commercialized in large scale as grain, bran and oil. Because it is an important source of protein, it plays a fundamental role in the food industry, both human and animal. Soybean monoculture in large areas has increased crop vulnerability to pathogens and insect pests, with significant consequences on grain production. Among such pest insects, stink bugs are considered a major pest of soybean crop, feeding directly on seeds, reducing yield and seed quality. Losses in crop yield due to stink bugs attack are greater than 30% and seed germination compromising can be greater than 50%. This study aimed to identify QTL (Quantitative Trait Loci), for stink bug resistance traits and agronomic traits using the multivariate multiple interval mapping. An F2 mapping population of 228 plants derived from a biparental cross between IAC-100 and CD-215 was used for genotyping. An F2:3 population was developed to evaluate eight stink bug resistant traits such as graining period, leaf retention, pod number per plant, percentage of pod damage, number of seeds, hundred seed weight, weight of healthy seeds and spotted seed weight. Other seven agronomic traits were evaluated such as number of days to flowering, plant height at flowering, number of days to maturity, plant height at maturity, lodging, agronomic value and grain yield. A total of 417 SNP (Single Nucleotide Polymorphism) markers, 61 SSR (Simple Sequence Repeat), 30 AFLP (Amplified Fragment Length Polymorphism) and 8 TRAP (Target Region Amplification Polymorphism) markers were mapped into 20 soybean linkage groups. The total map length was 2,814.82 cM with an average of 5.46 cM between markers. QTL mapping for those traits was performed using univariate (MIM) and multivariate (MT-MIM) multiple interval mapping, with main QTL effects estimates. MIM analysis identified a total of 60 QTL, through 13 soybean linkage groups, with 29 QTL related to stink bug resistant traits and 31 QTL related to agronomic traits. Phenotypic variation explained by QTL ranged from 14.27% for lodging to 65.45% for number of days to maturity. The traits were divided into nine groups for MT-MIM analysis considering their correlation coefficient. Twenty different genomic regions were identified showing a very high QTL clustering. For most of the traits phenotypic variation estimates for MT-MIM models were higher than MIM models, ranging from 27.98% to 65.30% for plant height at flowering and number of days to maturity, respectively. MT-MIM analysis showed 13 genomic regions controlling at least one of the evaluated traits which were not identified at MIM analysis. Comparison between identified QTL and QTL database at Soybase demonstrated that some QTL were similar to those described in different genetic background. However, 56 QTL detected in the pre...