Abstract
Genotypic data has been applied in multiple research fields, such as molecular biology, genetics, and breeding. However, due to various reasons, genotyping data inevitably contains a certain percentage of errors, which affects the reliability of gene localization. This study conducted quantitative trait locus (QTL) mapping for some yield related and disease resistant traits, using repeated genotyping data of two wheat RIL populations derived from Yangxiaomai × Zhongyou 9507 and Jingshuang 16 × Bainong 64, and non-erroneous data consisting of consistent genotypes between the two replications. Mapping results were compared with reported QTL for the corresponding traits. Then error correction methods implemented in software packages QTL IciMapping (EC), Genotype-Corrector (GC) and R/qtl were applied to these datasets, followed by QTL mapping. Simulation study was performed by randomly adding five levels of genotyping errors to investigate the effect of genotyping errors on QTL mapping and the efficiency of error correction methods. The results indicated that the impact of genotyping errors on QTL mapping was mainly reflected in the decrease in QTL detection power and accuracy of estimation for QTL positions and effects. The impact was increased with the increasing of errors. EC, GC, and R/qtl improved the mapping accuracy, but all of them cannot eliminate the negative effect of errors completely. Considering detection power, false discovery rate, estimation of QTL position and effect, correction accuracy, and running time, EC was better than the other two methods.