Novel Multilocus Measure of Linkage Disequilibrium to Estimate Past Effective Population Size

Table 1.

Results From Simulation of a 0.5-cM Chromosome Segment Containing 10 Markers, With N = 5000

Length of segment (cM) Markers in segment Expected CSH Observed CSH CV (%) t(generations ago) Nt 95% confidence interval
0.06 2 0.083 0.090 121 900 4540 3581–6098
0.11 3 0.043 0.046 147 450 4671 3363–7477
0.17 4 0.029 0.024 98 300 6028 4759–8184
0.22 5 0.022 0.023 80 225 4875 3949–6347
0.28 6 0.018 0.013 71 180 6679 5449–8610
0.33 7 0.015 0.011 72 150 6784 5411–9074
0.39 8 0.013 0.008 79 129 8028 6072–11,815
0.44 9 0.011 0.006 68 113 8806 6578–13,283
0.50 10 0.010 0.007 63 100 7377 5110–13,194
  • Results are from analysis of haplotypes after 30,000 simulated generations of breeding under the mutation drift model. The average heterozygosity of the markers was 0.39. The markers were equally spaced, with 0.06 cM between markers.

  • Calculated as 1/(4Nc + 1), where N = 5000 and c is the length of the segment in morgans.

  • The average results from 10 replicates. In all replicates, the minimum heterozygosity of the markers was 0.05, as this was the minimum heterozygosity of the markers in the Moffat et al. (2001) data set.

  • From pooled results over segments of the same length within a replicate and across replicates.

  • To calculate the 95% confidence intervals forNt, a 95% confidence interval for the observed CSH was calculated as average CSH ± 2SE. Then the upper and lower bounds of CSH were used to calculate the upper and lower bounds of the 95% confidence interval for Nt .

This Article

  1. Genome Res. 13: 635-643

Preprint Server