Skip to main content
. 2013 Jun 6;92(6):882–894. doi: 10.1016/j.ajhg.2013.04.023

Table 6.

Comparison of the Effect of Long-Range LD Regions of Varying Lengths on SMARTPCA and SPA

Method Inversion Length (SNPs) Euclidean Distance: 2nd[1st, 3rd] Quartile Distance: 2nd[1st, 3rd] Quartile (km) Relative to Length 0
SMARTPCA pruned 0 2.78 [1.67, 4.61] 247.1 [155.0, 378.3] 1
SMARTPCA pruned 410 3.18 [1.84, 5.30] 275.7 [162.7, 433.4] 1.12
SMARTPCA pruned 1,800 (whole) 6.98 [4.18, 11.9] 545.4 [362.6, 832.9] 2.21
SPA pruned 0 2.55 [1.56, 4.02] 226.4 [137.7, 336.2] 1
SPA pruned 450 2.60 [1.58, 4.03] 227.3 [139.5, 336.8] 1.00
SPA pruned 900 2.63 [1.58, 4.09] 230.3 [139.4, 344.6] 1.02
SPA pruned 1,800 (whole) 2.73 [1.61, 4.23] 240.6 [141.8, 355.7] 1.06

The genotypes of the Spanish samples from the chromosome 8 inversion were excised, trimmed to different lengths, and “transplanted” in the genotypes of the POPRES samples. “Inversion length” gives the number of SNPs in the transplanted inversion out of the entire 1,800 SNPs in the inversion region. Reported error measures are the same as in Table 1. “Relative to Length 0” gives, per method, the ratio between the median error (in km) and the result before the inversion was added.