(mean, animal, hys, y, gene content; missing value is -9999):
1 1 14825 -9999 1
1 2 5146 -9999 2
1 3 5386 -9999 1
1 4 47519 -9999 1
1 5 46051 -9999 1
1 6 43484 -9999 2
1 7 12719 -9999 1
1 8 27660 -9999 0
1 9 5681 -9999 0
1 10 21685 -9999 1
...
1 9999993 480244 99.7522132400075 -9999
1 9999994 461248 98.7729983384373 -9999
1 9999995 460282 100.339403361134 -9999
1 9999996 464308 101.923396214390 -9999
1 9999997 498089 99.8367733957329 -9999
1 9999998 494791 99.9258964576069 -9999
In blupf90:
# this models y= hys + u +e
# snp = mu + u +e
# correlated as in GCMTBLUP
# # no residual variance !! hence var(e) is fixed to a small value
DATAFILE
data
NUMBER_OF_TRAITS
2
NUMBER_OF_EFFECTS
3
OBSERVATION(S)
4 5
WEIGHT(S)
EFFECTS: POSITIONS_IN_DATAFILE NUMBER_OF_LEVELS TYPE_OF_EFFECT[EFFECT NESTED]
0 1 1 cross
3 0 500000 cross
2 2 10000000 cross
RANDOM_RESIDUAL VALUES
.95 0
0 .01
RANDOM_GROUP
3
RANDOM_TYPE
add_animal
FILE
ped
(CO)VARIANCES
.05 .11
.11 .5
OPTION missing -9999
In PEST, the parameter file is:
relationship
rel_for animal
disk
infile = 'ped'
input
animal
m_p
f_p
TRANSFORMATION
treated_as_missing
gc none -9999 none
ls none -9999 none
data
infile = 'data2'
disk
input
mean 1
animal 10000000
hys 500000
ls 0
gc 0
model
ls = hys animal
gc = mean animal
VE
0.95 0
0 0.01
VG
vg_for animal
0.05 0.11
0.11 0.50
solver
iod [stop = 0.001, relax = .7, max_iter = 2000
STAND_AVG_CHANGE
printout
outfile = 'vlistp20'
No comments:
Post a Comment