P0380 QTL Mapping in a Full-sib Family of Sugarcane Using Multiple Imputation

Carina O. Anoni , Luiz de Queiroz College of Agriculture, Department of Genetics, University of São Paulo, Piracicaba, Brazil
Maria M. Pastina , Luiz de Queiroz College of Agriculture, Department of Genetics, University of São Paulo, Piracicaba, Brazil
Rodrigo Gazaffi , Luiz de Queiroz College of Agriculture, Department of Genetics, University of São Paulo, Piracicaba, Brazil
Renato R. Silva , Luiz de Queiroz College of Agriculture, Department of Genetics, University of São Paulo, Piracicaba, Brazil
Marcelo Mollinari , Luiz de Queiroz College of Agriculture, Department of Genetics, University of São Paulo, Brazil, Piracicaba, Brazil
Thiago Gibbin Marconi , CBMEG - State University of Campinas, Brazil, Campinas, Brazil
Melina C Mancini , CBMEG - State University of Campinas, Brazil, Campinas, Brazil
Estela A Costa , CBMEG - State University of Campinas, Brazil, Campinas, Brazil
Luciana R. Pinto , Instituto Agronômico de Campinas, Centro de Cana, Ribeirão Preto, Brazil
Anete P. Souza , CBMEG - State University of Campinas, Brazil, Campinas, Brazil
Antonio Augusto F. Garcia , Luiz de Queiroz College of Agriculture, Department of Genetics, University of São Paulo, Piracicaba, Brazil
QTL mapping in sugarcane (Saccharum spp.) is important to understand the genetic architecture of quantitative traits that are important in breeding programs. However, the ocurrence of missing marker phenotypes is common and decrease the power to detect QTL and causes bias in estimates of locations and effects of QTL. Therefore, methods that include missing marker phenotypes should be considered. Our work was aimed at detecting QTL in a full-sib family of sugarcane via interval mapping method using the multiple imputation approach. The mapping population was composed of 220 individuals derived from a biparental cross between the Brazilian cultivars IAC95-3018 and IACSP93-3046. The evaluated traits related to yield were: fiber content (Fiber), sugar content (POL), cane yield (kg/plot), sugar yield (kg/plot). Ten imputation data sets were build using a 1-cM grid to infer pseudomarker genotype along the genetic linkage map. The QTL mapping on the data with imputed pseudomarker genotypes detected 57 QTLs; 14 QTL were obtained for Fiber; 19 for POL; 12 for cane yield and 12 for sugar yied. The LOD Score value and the R2 proportional to the average weight of all pseudomarker realizations at each grid position ranged from 3.82-7.52 and 6.49%-16.61%, respectively. In general, it was observed that additive and dominance effects were significant, with predominance of additive effects. The application of multiple imputation approach was successful in reducing the bias and increasing the power of QTL detection.