Multidimensional scaling for quality control analysis
Post-imputed and filtered markers were used to calculate a genome-wide pairwise identity-by-state (IBS) distance matrix for each family in TASSEL software using 1 - IBS followed by Multidimensional scaling (MDS) analysis\cite{Bradbury2007}. The first three principal coordinates in the multidimensional scaling of each family were graphically depicted using the R statistical software. The MDS analysis predicted the hidden population structure by separating four clusters of progeny vines relative to their corresponding parents and grandparents, and indicated vines that were self-pollinated, outcrosses, or mislabeled, which were removed from linkage mapping and GWAS analysis.  

Construction of genetic maps

Genetic maps were constructed in Lep-MAP3 (LM3) v. 0.2\cite{Rastas2017} using the VCF file of post-imputed and filtered markers as well as pedigree information for each family. The following LM3 modules and steps were used to construct the genetic maps: (1) ParentCall2 module of Lep-MAP3 was used to call parental genotypes; (2) the resulting output was filtered by using Filtering2 module (parameter dataTolerance = 1.00E-3 for F1 families and 1.00E-10 for the F2 family) and the markers were filtered out based on a χ2 (chi-squared) test (testing if the allele ratio is significantly deviated from the expected mendelian ratio, P <0.001) or monomorphism; (3) SeparateChromosomes2 module was used to identify linkage groups using logarithm of odds (LOD) score limit ranging from none to 20 for the individual family (Table 1); (4) Finally, OrderMarkers2 module was used to compute the parental genetic distances (sex specific for the F1 families and sex averaged for the F2 family) of the markers in the linkage groups using 20 iterations per group. Correlation plots of genetic and physical distances of individual markers per chromosome in each family were plotted to evaluate the consistency of the maps, genome organization and structural variation.