Table 5 . Comparisons of ibd inference using using ibd_haplo and using fastIBD.

Program:		True-positive ibd inference				False-positive ibd inference
Input	Scoring	LD level γ				LD level γ
		0.0	0.05	0.1	1.0	0.0	0.05	0.1	1.0
ibd_haplo:	All ibd	69.3	74.5	77.0	80.9	17.2	8.5	3.8	0.1
Hap data	Btw ibd	69.8	74.5	77.5	81.3
ibd_haplo:	All ibd	53.4	57.6	59.8	61.7	12.1	6.5	3.4	0.1
Gen data	Btw ibd	46.1	49.7	52.3	53.3
fastIBD:	Strict	31.4	41.2	46.4	56.9	0.6	0.7	0.8	1.0
100 base	Loose	56.3	64.6	68.9	82.9	2.2	2.3	2.3	4.0
fastIBD:	Strict	41.5	51.4	56.9	71.7	0.7	0.9	0.9	1.5
1000 base	Loose	66.5	73.3	77.6	90.4	3.4	3.1	3.1	8.9

Results are shown at decreasing levels of LD (that is, increasing γ), and are tabulated as percentages over the 6913 markers over all 500 pairs of individuals. The ibd haplo program was run for both haplotypic (Hap.) and genotypic data (Gen.). For scoring ibd all states involving any ibd were scored (All ibd), as in Table 3, and then also only between-individual ibd was scored (Btw ibd), since fastIBD does not seek within-individual ibd. The fastIBD program was run using only the same 100 individuals for the BEAGLE imputation and phasing step as used in ibd haplo, and then also using an additional 900 individuals for a base sample of 1000. The ibd was scored using a strict (10⁻¹⁰) threshold and a looser threshold (10⁻⁶).