Supplementary data of "Improving multi-breed genomic prediction by modeling heterogeneous genetic (co)variance blockwise accounting for linkage disequilibrium"

The compressed file includes all data files necessary to reproduce the results of our study "Improving multi-breed genomic selection by modeling heterogeneous genetic (co)variance with a blockwise approach accounting for linkage disequilibrium", along with random number seeds for replicating associated sampling or simulation processes.The data contained in the folder Keller2022 comes from Keller et al., which consisted of three bean breeding panels: climbing bean panel (VEC), Andean diversity panel (ADP), and the elite Andean breeding panel (VEF), whereas the latter two were bush type bean. Two traits 100-seed weight (100SdW) and yield (Yield) were analysed. The best linear unbiased estimators (BLUEs) for each trait calculated by Keller et al. (2022) were used as the response variable in our study. The numbers of lines with trait records for each panel were from 344 to 587. The filtrated genotype data with 14,913 SNPs located on 11 chromosomes were used in our study.For a detailed introduction to the data, please refer to: https://doi.org/10.7910/DVN/RLAWYNThe data contained in the folder Xie2021 comes from Xie et al. The populations including 228 Landrace (LL) and 641 Yorkshire (YY). Two traits marbling score (MS) and proportion of fat areas in the image (PFAI) were analysed. All individuals in the dataset had both phenotypes and genotypes. After removing non-autosomal SNPs and those with a minor allele frequency (MAF) less than 0.01, a total number of 37,304 SNPs were remained for subsequent analysis.For a detailed introduction to the data, please refer to: https://doi.org/10.6084/m9.figshare.14274785The code and program files related to the dataset can be accessed at the following GitHub website link:https://github.com/CAU-TeamLiuJF/GP-cross-validation