Table 1.
Experimental variables in pedigree-based and genomic selection analyses for BCWD resistance.
Variable | PED modela | Progeny testingb | ssGBLUP and wssGBLUPc | BayesB and BayesCd | ||
---|---|---|---|---|---|---|
Training | Validation | Training | Validation | |||
Number of families | 71 | 31 | 71 | 10 | 10 | 10 |
Offspring per family | 39–80 | 38–122 | 39–80 | 2–11 | 39–80 | 2–11 |
Genotyped fish | Nae | Na | 658 | 53 | 583 | 53 |
Phenotyped fish | 4492 | 1913 | 4492 | Na | 583 | Na |
Pedigree records | 4757 | Na | 4757 | Na | Na | Na |
Progeny tested breedersf | Na | 53 | Na | 53 | Na | 53 |
Pedigree-based model (PED) fit BCWD records from 2005 families.
The validation fish or potential breeders were mated to generate 31 progeny testing full-sib families.
The single-step GBLUP (ssGBLUP) and weighted ssGBLUP (wssGBLUP) methods used in training models analysis all fish that had genotype and phenotype records and had any type of pedigree relationship (parents, full-sibs, half-sibs, etc.).
The Bayesian methods BayesB and BayesC used in training model analysis only those fish that had matched genotype and phenotype records without missing data.
Na indicates either non-available or non-needed data cell.
Progeny tested breeders are fish from 2005 families that were used as breeders to generate 2007 families.