Skip to main content

Table 2 Variance components (with SE) estimated in Large White sows for litter size (TNB) and residual variance of litter size (varTNB) using a Double Hierarchical GLM (DHGLM) and for mean litter size per sow (meanTNB) and log-transformed variance of the TNB per sow (log (var (TNB)) using conventional bivariate analysis

From: Genome-wide association study reveals novel loci for litter size and its variability in a Large White pig population

 

DHGLM

Conventional

Estimates

TNB

varTNB

meanTNB

log(var(TNB))

Additive genetic variance

1.18 (0.04)

0.03 (0.003)

1.23 (0.04)

0.04 (0.004)

Permanent sow variance

0.69 (0.02)

0.15 (0.004)

−a

−a

Residual variance

6.5 (0.02)b

1.88 (0.01)b

11.8 (0.15)

3.78 (0.03)

Heritability

0.14 (0.003)b

0.006c (0.0008)b

0.09 (0.001)

0.01 (0.006)

GCVSDe

d

 

0.087 (0.004)b

  
  1. aThe conventional analysis has no permanent sow effect, since there is only a single observation per sow
  2. bStandard errors obtained based on calculations from Mulder et al. [82]
  3. cHeritability estimated at the level of squared phenotype: h 2v  = σ 2av /(2σ 4P  + 3(σ 2av  + σ 2pev )) [70]
  4. dGenetic coefficient of variation at residual standard deviation level, i.e. the genetic standard deviation in residual standard deviation divided by the mean residual standard deviation of the trait: \( {\mathrm{GCV}}_{\mathrm{SDe}}=\frac{\sigma_{\mathrm{av}}\left({\sigma}_{\mathrm{e}}\right)}{\overline{\sigma_{\mathrm{e}}}}\approx {\scriptscriptstyle \frac{1}{2}}{\sigma}_{\mathrm{av}} \) [1]