A multi-trait Bayesian method for mapping QTL and genomic prediction

Table 5 Posterior mean number of SNPs^a in each distribution for milk production traits from BayesMV or BayesR

Reference	Distribution^b,c	BayesR	BayesMV
Hol_Jer	Unassociated	–	627,911
	LC1₁_LC2₁_LC3₁	620,515	1
	LC1₁_LC2₁_LC3_2–4	3504	11
	LC1₁_LC2_2–4_LC3₁	2994	4
	LC1_2–4_LC2₁_LC3₁	4913	64
	LC1₁_LC2_2–4_LC3_2–4	21	47
	LC1_2–4_LC2₁_LC3_2–4	29	685
	LC1_2–4_LC2_2–4_LC3₁	25	218
	LC1_2–4_LC2_2–4_LC3_2–4	1	3062
Holstein	Unassociated	–	628,451
	LC1₁_LC2₁_LC3₁	621,268	0
	LC1₁_LC2₁_LC3_2–4	2817	2
	LC1₁_LC2_2–4_LC3₁	3110	4
	LC1_2–4_LC2₁_LC3₁	4743	12
	LC1₁_LC2_2–4_LC3_2–4	17	50
	LC1_2–4_LC2₁_LC3_2–4	22	124
	LC1_2–4_LC2_2–4_LC3₁	25	234
	LC1_2–4_LC2_2–4_LC3_2–4	0	3124
Jersey	Unassociated	–	630,779
	LC1₁_LC2₁_LC3₁	624,314	0
	LC1₁_LC2₁_LC3_2–4	1957	3
	LC1₁_LC2_2–4_LC3₁	1366	3
	LC1_2–4_LC2₁_LC3₁	4335	1
	LC1₁_LC2_2–4_LC3_2–4	6	101
	LC1_2–4_LC2₁_LC3_2–4	14	28
	LC1_2–4_LC2_2–4_LC3₁	10	30
	LC1_2–4_LC2_2–4_LC3_2–4	0	1057

^aThe posterior mean number of unassociated SNPs from BayesMV is shown with the joint probability of a non-zero effect on one or more traits. Joint probabilities are the product of posterior probabilities (p and q)
^bTraits are three linear combinations (LC1, LC2, LC3) of fat, milk and protein yield
^cSubscripts indicate distributions 1 to 4, each explaining 0, 0.0001, 0.001 or 0.01 of the genetic variance

ISSN: 1297-9686