Evaluation and optimization of a genome-wide selection method based on cotton fiber quality traits

• 1、College of Agriculture, Hebei Agricultural University, Baoding 071000

Abstract：The application of genomic selection in cotton is still in the development stage and many factors affecting genomic selection in cotton need to be further study. Based on previous studies, the whole-genome prediction of 4 fiber quality traits in 1224 upland cotton samples was conducted by using 12 parameter and non-parametric prediction models, and the effects of marker density, population size and population structure on the prediction accuracy were investigated. The results show that the LASSO method is the most robust among the 12 methods, and the prediction effect of the parametric method is higher than that of the non-parametric method. The prediction accuracy of the four fiber quality traits increased gradually with the increase of the number of markers and the size of the training population until reaching the plateau. It was found that the closer the genetic relationship between the training population and the prediction population, the higher the prediction accuracy. The results of this study will further accelerate the research process of the whole genome selection of cotton and provide some theoretical basis and reference for the practical breeding of cotton.

Keywords： Genome selection Fiber quality traits Accuracy of prediction