Skip to main content

Table 4 Computing time for the imputation of high-density SNP on chromosomes 1, 15 and 25

From: Imputation of high-density genotypes in the Fleckvieh cattle population

Number of animals in reference and validation population

Chr

Beagle

MaCH a

findhap.f90b

Minimac c

50 / 747

BTA1

2.67 h

1.30 h (0.03 h / 0.30 h / 0.97 h)

0.07 h

0.17 h (0.03 h / 0.07 h / 0.07 h)

BTA15

1.18 h

0.68 h (0.02 h / 0.14 h / 0.52 h)

0.04 h

0.09 h (0.02 h / 0.04 h / 0.03 h)

BTA25

0.67 h

0.37 h (0.01 h / 0.08 h / 0.28 h)

0.04 h

0.05 h (0.01 h / 0.02 h / 0.02 h)

100 / 697

BTA1

3.93 h

5.01 h (0.08 h / 1.11 h / 3.82 h)

0.07 h

0.27 h (0.08 h / 0.06 h / 0.13 h)

BTA15

2.48 h

2.72 h (0.05 h / 0.55 h / 2.12 h)

0.05 h

0.15 h (0.05 h / 0.03 h / 0.07 h)

BTA25

1.33 h

1.48 h (0.03 h / 0.32 h / 1.13 h)

0.04 h

0.09 h (0.03 h / 0.02 h / 0.04 h)

200 / 597

BTA1

4.49 h

18.92 h (0.20 h / 4.31 h / 14.41 h)

0.07 h

0.48 h (0.20 h / 0.05 h / 0.23 h)

BTA15

2.87 h

10.06 h (0.11 h / 2.22 h / 7.73 h)

0.05 h

0.27 h (0.11 h / 0.03 h / 0.13 h)

BTA25

1.38 h

5.76 h (0.06 h / 1.24 h / 4.45 h)

0.04 h

0.14 h (0.06 h / 0.01 h / 0.07 h)

400 / 397

BTA1

3.73 h

81.23 h (0.44 h / 21.97 h / 58.82 h)

0.07 h

1.1 h (0.44 h / 0.03 h / 0.63 h)

BTA15

2.45 h

40.16 h (0.21 h / 10.52 h / 29.43 h)

0.05 h

0.56 h (0.21 h / 0.02 h / 0.33 h)

BTA25

1.37 h

28.30 h (0.11 h / 5.98 h / 22.21 h)

0.04 h

0.30 h (0.11 h / 0.01 h / 0.18 h)

  1. The number of imputed SNP was 36 599, 20 145 and 10 981 for chromosomes 1, 15 and 25, respectively. Computing was performed on an Intel Xeon 2.13 Ghz processor.
  2. a The entire computing time for MaCH can be partitioned into three separate steps (in parentheses): pre-phasing of the reference population with Beagle, inference of tuning parameters based on 200 randomly selected animals of the validation population and actual genotype imputation with MaCH.
  3. bfindhap.f90 was run exploiting the multi-threading option.
  4. c The entire computing time for Minimac can be partitioned into three separate steps (in parentheses): pre-phasing of the reference population with Beagle, pre-phasing of the validation population with Beagle and actual genotype imputation with Minimac.