Skip to main content

Table 1 Results of a general linear model testing factors affecting of allele frequency estimation

From: Fish scales and SNP chips: SNP genotyping and allele frequency estimation in individual and pooled DNA from historical samples of Atlantic salmon (Salmo salar)

Pooled individual number Model terms Parameter estimate S.E. t-value P-value
514 Individuals Minor Allele Frequency 0.0211 0.00180 11.75 < 0.001
(N LOCI = 4642) GenTrain Score −0.0226 0.00196 −11.49 < 0.001
  Locus Classification:     
  Mono −0.0149 0.00094 −15.74 < 0.001
  MSV-3 −0.0045 0.00102 −4.401 < 0.001
20 individuals Minor Allele Frequency 0.0199 0.00352 5.65 < 0.001
(N LOCI = 4571) GenTrain Score −0.0227 0.00409 −5.55 < 0.001
  Locus Classification:     
  Mono −0.0150 0.00185 −8.14 < 0.001
  MSV-3 0.0063 0.00186 3.37 < 0.001
50 individuals Minor Allele Frequency 0.0093 0.00346 2.68 0.00735
(N LOCI = 4681) GenTrain Score −0.0302 0.00407 −7.43 < 0.001
  Locus Classification:     
  Mono −0.0145 0.00182 −7.96 < 0.001
  MSV-3 −0.0035 0.00196 −1.79 0.0730
  1. The effects of minor allele frequency, GenTrain score and locus classification on the accuracy of allele frequency estimates were tested based on the mean difference between empirical and estimated allele frequencies per locus for all individuals. Estimated allele frequencies were estimated from clusters of all individuals (N = 514) and random subsets of 20 and 50 individuals. The mean difference in allele frequency was used as a dependent variable with a Gaussian error structure. S.E. is the standard error, t-value is the t statistic value and P-value is the corresponding significance of the model term.