TABLE 1

Phasing accuracy—simulated data

Familial information (steps I and II)+ Population information (step III)Overall
TypePhased Het SNPsSwitchesPhased Het SNPsSwitches FPMSwitches LHCMSwitches FPMSwitches LHCM
None (sires, <6 sons)0 (0.0%)a0.0 (0.0%)687 (100.0%)8.2 (1.2%)3.7 (0.5%)8.2 (1.2%)3.7 (0.5%)
0 (0.0%)b0.0 (0.0%)3525 (100.0%)24.9 (0.7%)4.0 (0.1%)24.9 (0.7%)4.0 (0.1%)
I (sons, <5 sibs)334 (49.3%)0.0 (0.0%)343 (50.7%)2.3 (0.7%)1.8 (0.5%)2.3 (0.3%)1.8 (0.3%)
1725 (49.5%)0.0 (0.0%)1763 (50.5%)6.1 (0.3%)3.6 (0.2%)6.1 (0.2%)3.6 (0.1%)
IIa (sires, ≥6 sons)673 (98.3%)0.1 (0.0%)12 (1.7%)0.0 (0.0%)0.0 (0.0%)0.1 (0.0%)0.1 (0.0%)
3462 (98.3%)0.1 (0.0%)60 (1.7%)0.1 (0.2%)0.0 (0.0%)0.2 (0.0%)0.2 (0.0%)
IIa (sires, >10 sons)684 (99.9%)0.0 (0.0%)0.3 (0.1%)0.0 (0.0%)0.0 (0.0%)0.0 (0.0%)0.0 (0.0%)
3525 (99.9%)0.0 (0.0%)1.4 (0.1%)0.0 (0.0%)0.0 (0.0%)0.0 (0.0%)0.0 (0.0%)
I+IIa (sons, ≥5 sibs)334 (49.3%)0.0 (0.0%)343 (50.7%)0.1 (0.0%)0.1 (0.0%)0.1 (0.0%)0.1 (0.0%)
1773 (50.1%)0.0 (0.0%)1763 (49.9%)0.3 (0.0%)0.2 (0.0%)0.3 (0.0%)0.2 (0.0%)
I+IIa (sons, ≥10 sibs)334 (49.3%)0.0 (0.0%)343 (50.7%)0.1 (0.0%)0.1 (0.0%)0.1 (0.0%)0.1 (0.0%)
1724 (49.4%)0.0 (0.0%)1764 (50.6%)0.2 (0.0%)0.2 (0.0%)0.2 (0.0%)0.2 (0.0%)
I+IIa+IIb (sons, ≥5 sibs)617 (91.1%)3.0 (0.5%)60 (8.9%)0.1 (0.2%)0.1 (0.2%)3.1 (0.5%)3.1 (0.5%)
3237 (92.9%)13.1 (0.4%)249 (7.1%)0.2 (0.1%)0.1 (0.0%)13.2 (0.4%)13.1 (0.4%)
I+IIa+ IIb (sons, ≥10 sibs)617 (91.1%)0.2 (0.0%)60 (8.9%)0.1 (0.2%)0.0 (0.0%)0.3 (0.0%)0.2 (0.0%)
3240 (92.9%)0.5 (0.0%)248 (7.1%)0.1 (0.0%)0.0 (0.0%)0.7 (0.0%)0.6 (0.0%)
  • Information used for phasing: I, use of Mendelian segregation rules to reconstruct the offspring's phase; IIa, use of linkage information to reconstruct the parental phase (marker alleles that tend to be cotransmitted reside on the same homolog); IIb, use of linkage information to reconstruct the offspring's phase (a chromosome segment flanked by two alleles originating from the same parental homolog is assumed to completely derive from that homolog unless the probability of double cross-over is > 0.999; III, use of population-wide linkage disequilibrium information.

  • a Low-density map (2364 markers).

  • b High-density map (12,113 markers). Of the markers corresponding to 1685 and 8623 markers for the low- and high-density maps, respectively, 71% were homozygous and considered phased de facto. The numbers shown pertain to remaining heterozygous markers only. In all instances, the FPM and LHCM models were trained on a haploid set of base haplotypes obtained by extracting type I information for all sons and type IIa information for sires with 6 or more sons.