Skip to main content

Table 2 Tests for QTL effects and corresponding empirical thresholds for 5% type 1 error based on 1000 permutations

From: Detecting parent of origin and dominant QTL in a two-generation commercial poultry pedigree using variance component methodology

Test

QTL in Model

QTL effect tested for

Bodyweight

Conformation score

 

alternative (H1)

null (H0)

 

*LRT (5%)

*LRT (5%)

Add

add (2)

null (1)

additive

5.74

4.53

Addom

add + dom (3)

null (1)

additive + dominant

6.98

5.84

pat + mat

pat + mat (4)

null (1)

paternal + maternal

3.05

2.94

Pat

pat (5)

null (1)

paternal

7.16

6.6

Mat

mat (6)

null (1)

maternal

5.38

4.54

Dom

add + dom (3)

add (2)

dominant

4.80

5.12

Imp

pat + mat (4)

add (2)

parent of origin

3.18

3.43

**patvfull

pat + mat (4)

pat (5)

maternally expressed

4.14

4.32

**matvfull

pat + mat (4)

mat (6)

paternally expressed

4.5

3.58

  1. * LRT is the chromosome-wise empirical threshold for 5% type 1 error rate for test statistic (twice the difference between log for the alternative and null models), estimated by 1000 iterations.
  2. ** For example, if the test of patvfull is significant the model incorporating paternal and maternal QTL is explaining more variation than the paternal QTL indicating some level of maternal expression. If there is no significant difference between the pat + mat model and mat model the maternal QTL is explaining all the variation.