Table 4 Performance of different genotype callers for calling diploid genotypes from nucleotide data at triallelic sites
MethodMean CoverageCorrect-Call Rate (mean ± 2 SEM)Correct-Call Rate among Called Genotypes (mean ± 2 SEM)
HGC100.97 ± 0.000370.97 ± 0.00037
GATK100.97 ± 0.000320.98 ± 0.00031
Samtools100.96 ± 0.000400.96 ± 0.00040
BGC100.78 ± 0.000990.79 ± 0.00099
HGC150.99 ± 0.000220.99 ± 0.00022
GATK151.00 ± 0.000131.00 ± 0.00013
Samtools150.96 ± 0.000400.96 ± 0.00040
BGC150.80 ± 0.000880.81 ± 0.00087
HGC201.00 ± 0.000141.00 ± 0.00014
GATK201.00 ± 0.000071.00 ± 0.00007
Samtools200.97 ± 0.000310.97 ± 0.00031
BGC200.80 ± 0.000840.81 ± 0.00083
HGC301.00 ± 0.000051.00 ± 0.00005
GATK301.00 ± 0.000021.00 ± 0.00002
Samtools300.99 ± 0.000150.99 ± 0.00015
BGC300.81 ± 0.000800.81 ± 0.00080
  • Allele frequencies p, q, and r are 0.7, 0.2, and 0.1, and the population is in Hardy–Weinberg equilibrium. Correct-call rate and that among called genotypes are calculated among N = 100 individuals and those with called genotypes, respectively. Error rate ε = 0.01. Results are based on a total of 10,000 simulation replications for each parameter set.