Page 295 - Linear Models for the Prediction of Animal Breeding Values 3rd Edition

The pedigree file is created and ordered in such a manner that contributions to the
diagonal and RHS of an animal from the pedigree due to the number of parents
known (see type 1 record below) and from progeny accounting for whether mate is
known (type 2 record), can be accumulated while processing the animal. Thus, ini-
tially, a pedigree file is created consisting of two types of records:
1. Type 1 record for all animals in the data comprising the animal identity, record
type, and sire and dam identities.
2. Type 2 record for each parent in the data comprising the parent identity, record
type, identities for progeny and other parent (mate) if known. The type 2 records are
used to adjust the contribution of the progeny to each parent for the mate’s breeding
value when solving for animal solutions.
The pedigree file is sorted by animal and record type. The sorted pedigree file for
the example data is given below.

Animal Code Sire or progeny Dam or mate
1 1 0 0
1 2 4 0
1 2 6 2
2 1 0 0
2 2 5 3
2 2 6 1
3 1 0 0
3 2 5 2
3 2 8 6
4 1 1 0
4 2 7 5
5 1 3 2
5 2 7 4
6 1 1 2
6 2 8 3
7 1 4 5
8 1 3 6

Second, a data file is set up consisting of animal identity, fixed effects, covariates
and traits. If there is a major fixed effect with many levels, two data files need to be
set up: one sorted by the major fixed effects such as herd or HYS (file A), to be used
when solving for the major fixed effect; and the other sorted by animal identity (file B),
to be used to solve for animal solutions. Assuming sex effect to be the major fixed
effect in the example data, the data sorted by sex are as follows:

Weaning weight
Calf Sex gain (kg)
4 Male 4.5
7 Male 3.5
8 Male 5.0
5 Female 2.9
6 Female 3.9

Solving Linear Equations 279