Supplemental Material for Schlather et al., 2019

File S1: Additional method section on setting a target coverage.

File S2: Additional method section on the graphical representation of haplotype blocks.

File S3: Additional method section - toy example for calculation of bEHH.

File S4: Script to simulate a diallel design in MoBPS

Table S1: Proportion of variance explained between the full SNP-dataset (X), a SNP-subset (Xs) and the block dataset (Z). For comparability the number of parameters in Xs and Z were chosen equally using traditionally heritability estimation as in de los Campos (2017).

Table S2: Influence of the window size on the haplotype library for chromosome 1 in the KE DH-lines with a target coverage of 95%.

Table S3: Influence of the weighting of block length (wl) and number of haplotypes (wn) on the haplotype library for chromosome 1 in the KE DH-lines with a target coverage of 95%.

Table S4: Influence of using the extended-block-identification on the haplotype library in dependency of the parameter t of the extended-block-identification-step for chromosome 1 in the KE DH-lines with a target coverage of 95%.

Figure S1: Dataset for the toy example used in File S3.

Figure S2: Comparison of the block structure and a bifurcation plot (Sabeti et al., 2002, Gautier and Vitalis, 2012) according SNP 10,000 on chromosome 1 in the KE DH-lines.

Figure S3: Comparison of the block structure for MCMB=1000 (A), MCMB=5000 (B), MCMB=20,000 (C) for the first 20,000 SNPs of chromosome 1 in the KE DH-lines.