ASReml - Singularity appearing in simple animal model

Sev — Tue, 31 Jan 2012 08:53:13 GMT

Hi. I am experiencing difficulty with running an animal model for a collared flycatcher dataset. I actually wanted to run random regression models but have been unable to run even a simple animal model because of a singularity being reported in the model. The issue arises when both additive genetic and permanent environment effects are included as random effects in the model - when either of these is omitted the model seems to run fine. However, a singularity is reported when both are included. The component of the permanent environment effect is consistently estimated as being 10% of that of the residual variance.

I have tried modelling different traits, using alternative versions of the pedigree and a much smaller version of the dataset without success - always the result is the same, with the model reporting a singularity on the first run and the permanent environment component being estimated as 10% of the residual variance. I also used a colleague's pedigree and dataset, which he said were working fine for him, but the same problem has arisen.

I can see that I must be doing something wrong but I have run out of ideas and so I am posting this in the hope that someone with a much greater understanding of the program might be able to offer a potential solution.

Thanks

ASReml - RE: at(grp):member

david.butler — Tue, 31 Jan 2012 05:16:42 GMT

Hi Ian,

I just tried a 'dummy' example with 200 families and 20 lines/family and the construct at(family):line worked ok. In fact, I think the limit is now 9999 for simple components. I wonder if I could trouble you for a suitably sanitized version of your job so I can get to the bottom of the problem.

Dave.

ASReml - Modeling block-diagonal residual matrix in RR models

kon — Mon, 30 Jan 2012 22:17:44 GMT

Hello there,

I am trying to run a random regression model with block-diagonal residual matrix.
I cannot figure out the syntax for this situation.
Can anybody help me with the syntax? Attached is my *.as file. Thanks in advance.

Kon

Run1 milk fat protein 3 lactations test data
htd1 9 !m0
htd2 6 !m0
htd3 5 !m0
animal !P
p2 12
age1 !m0
age2 !m0
age3 !m0
testNo 10 !m0
dim1 !m0
dim2 !m0
dim3 !m0
milk1 !m0
fat1 !m0
prot1 !m0
milk2 !m0
fat2 !m0
prot2 !m0
milk3 !m0
fat3 !m0
prot3 !m0
scs1 !m0
scs2 !m0
scs3 !m0

mfp.ped !MAKE
mfp.dat !MVINCLUDE !AISING !ASUV !BLUP
milk1 fat1 prot1 milk2 fat2 prot2 milk3 fat3 prot3 ~ Trait,
at(Tr,1).htd1 at(Tr,2).htd1 at(Tr,3).htd1,
at(Tr,4).htd2 at(Tr,5).htd2 at(Tr,6).htd2,
at(Tr,7).htd3 at(Tr,.htd3 at(Tr,9).htd3,
!r ![at(Tr,1).leg(dim1,2).animal at(Tr,2).leg(dim1,2).animal at(Tr,3).leg(dim1,2).animal,
at(Tr,4).leg(dim2,2).animal at(Tr,5).leg(dim2,2).animal at(Tr,6).leg(dim2,2).animal,
at(Tr,7).leg(dim3,2).animal at(Tr,.leg(dim3,2).animal at(Tr,9).leg(dim3,2).animal !],
![at(Tr,1).leg(dim1,2).p2 at(Tr,2).leg(dim1,2).p2 at(Tr,3).leg(dim1,2).p2,
at(Tr,4).leg(dim2,2).p2 at(Tr,5).leg(dim2,2).p2 at(Tr,6).leg(dim2,2).p2,
at(Tr,7).leg(dim3,2).p2 at(Tr,.leg(dim3,2).p2 at(Tr,9).leg(dim3,2).p2 !] ! mv
3 2 2 !STEP 0.01 ASMV 10
#12 !S2==1

#90 0 DIAG !+90
# 0.657908 0.637942 0.771447 0.657908 0.637942 0.771447 0.657908 0.637942 0.771447
# 0.422727 0.453957 0.526399 0.422727 0.453957 0.526399 0.422727 0.453957 0.526399
# 0.386030 0.402864 0.442875 0.386030 0.402864 0.442875 0.386030 0.402864 0.442875
# 0.320358 0.322828 0.377327 0.320358 0.322828 0.377327 0.320358 0.322828 0.377327
# 0.266839 0.270653 0.303710 0.266839 0.270653 0.303710 0.266839 0.270653 0.303710
# 0.226540 0.221478 0.253297 0.226540 0.221478 0.253297 0.226540 0.221478 0.253297
# 0.230738 0.204920 0.216980 0.230738 0.204920 0.216980 0.230738 0.204920 0.216980
# 0.213249 0.191251 0.202024 0.213249 0.191251 0.202024 0.213249 0.191251 0.202024
# 0.192626 0.164194 0.187223 0.192626 0.164194 0.187223 0.192626 0.164194 0.187223
# 0.188050 0.142975 0.141640 0.188050 0.142975 0.141640 0.188050 0.142975 0.141640

How to model this matrix?

Period I lact II lact II lact
(Test No)

39.31921 22.01876 18.08683 47.19110 18.87644 16.51689 49.50758 14.85227 13.36705
1 22.01876 23.51595 11.28766 18.87644 28.82707 9.22466 14.85227 30.13304 7.53326
18.08683 11.28766 14.03567 16.51689 9.22466 16.21727 13.36705 7.53326 16.82260

27.52344 15.41313 12.66078 33.03377 13.21351 11.56182 34.65530 10.39659 9.35693
2 15.41313 16.46117 7.90136 13.21351 20.17895 6.45726 10.39659 21.09313 5.27328
12.66078 7.90136 9.82497 11.56182 6.45726 11.35209 9.35693 5.27328 11.77582

26.34387 14.75257 12.11818 31.61804 12.64721 11.06631 33.17007 9.95102 8.95592
3 14.75257 15.75569 7.56273 12.64721 19.31414 6.18052 9.95102 20.18914 5.04728
12.11818 7.56273 9.40390 11.06631 6.18052 10.86557 8.95592 5.04728 11.27114

25.16429 14.09200 11.57557 30.20230 12.08092 10.57081 31.68485 9.50545 8.55491
4 14.09200 15.05021 7.22410 12.08092 18.44932 5.90378 9.50545 19.28514 4.82129
11.57557 7.22410 8.98283 10.57081 5.90378 10.37906 8.55491 4.82129 10.76647

23.98472 13.43144 11.03297 28.78657 11.51463 10.07530 30.19962 9.05989 8.15390
5 13.43144 14.34473 6.88547 11.51463 17.58451 5.62704 9.05989 18.38115 4.59529
11.03297 6.88547 8.56176 10.07530 5.62704 9.89254 8.15390 4.59529 10.26179

22.80514 12.77088 10.49036 27.37084 10.94834 9.57979 28.71439 8.61432 7.75289
6 12.77088 13.63925 6.54684 10.94834 16.71970 5.35030 8.61432 17.47716 4.36929
10.49036 6.54684 8.14069 9.57979 5.35030 9.40602 7.75289 4.36929 9.75711

21.62556 12.11032 9.94776 25.95511 10.38204 9.08429 27.22917 8.16875 7.35187
7 12.11032 12.93377 6.20821 10.38204 15.85489 5.07356 8.16875 16.57317 4.14329
9.94776 6.20821 7.71962 9.08429 5.07356 8.91950 7.35187 4.14329 9.25243

20.44599 11.44975 9.40515 24.53937 9.81575 8.58878 25.74394 7.72318 6.95086
8 11.44975 12.22829 5.86958 9.81575 14.99008 4.79682 7.72318 15.66918 3.91729
9.40515 5.86958 7.29855 8.58878 4.79682 8.43298 6.95086 3.91729 8.74775

19.26641 10.78919 8.86255 23.12364 9.24946 8.09327 24.25871 7.27761 6.54985
9 10.78919 11.52282 5.53095 9.24946 14.12526 4.52008 7.27761 14.76519 3.69130
8.86255 5.53095 6.87748 8.09327 4.52008 7.94646 6.54985 3.69130 8.24308

18.08683 10.12863 8.31994 21.70791 8.68316 7.59777 22.77348 6.83205 6.14884
10 10.12863 10.81734 5.19232 8.68316 13.26045 4.24334 6.83205 13.86120 3.46530
8.31994 5.19232 6.45641 7.59777 4.24334 7.45995 6.14884 3.46530 7.73840

at(Tr,1).leg(dim1,2).animal 2
27 0 US
100.51217
-1.04420 18.17910
0.60234 -0.81780 4.93379
0.74988 -0.00779 0.00449 38.70681
-0.00779 0.13563 -0.00610 -0.40212 7.00069
0.00449 -0.00610 0.03681 0.23196 -0.31493 1.89998
1.29391 -0.01344 0.00775 0.91162 -0.00947 0.00546 29.59022
-0.01344 0.23402 -0.01053 -0.00947 0.16488 -0.00742 -0.30741 5.35183
0.00775 -0.01053 0.06351 0.00546 -0.00742 0.04475 0.17733 -0.24076 1.45248
1.16158 -0.01207 0.00696 0.61755 -0.00642 0.00370 1.02925 -0.01069 0.00617 124.65933
-0.01207 0.21009 -0.00945 -0.00642 0.11169 -0.00502 -0.01069 0.18615 -0.00837 -1.29506 22.54646
0.00696 -0.00945 0.05702 0.00370 -0.00502 0.03031 0.00617 -0.00837 0.05052 0.74705 -1.01427 6.11909
0.58814 -0.00611 0.00352 1.20569 -0.01253 0.00723 0.72047 -0.00748 0.00432 0.92632 -0.00962 0.00555 52.00552
-0.00611 0.10637 -0.00479 -0.01253 0.21807 -0.00981 -0.00748 0.13031 -0.00586 -0.00962 0.16754 -0.00754 -0.54028 9.40596
0.00352 -0.00479 0.02887 0.00723 -0.00981 0.05918 0.00432 -0.00586 0.03537 0.00555 -0.00754 0.04547 0.31166 -0.42313 2.55277
0.98513 -0.01023 0.00590 0.79399 -0.00825 0.00476 1.16158 -0.01207 0.00696 1.32332 -0.01375 0.00793 1.07336 -0.01115 0.00643 36.67043
-0.01023 0.17818 -0.00802 -0.00825 0.14360 -0.00646 -0.01207 0.21009 -0.00945 -0.01375 0.23934 -0.01077 -0.01115 0.19413 -0.00873 -0.38096 6.63238
0.00590 -0.00802 0.04836 0.00476 -0.00646 0.03897 0.00696 -0.00945 0.05702 0.00793 -0.01077 0.06496 0.00643 -0.00873 0.05269 0.21976 -0.29836 1.80002
1.02925 -0.01069 0.00617 0.51462 -0.00535 0.00308 0.92632 -0.00962 0.00555 1.26450 -0.01314 0.00758 0.74988 -0.00779 0.00449 1.14687 -0.01191 0.00687 137.75070
-0.01069 0.18615 -0.00837 -0.00535 0.09308 -0.00419 -0.00962 0.16754 -0.00754 -0.01314 0.22870 -0.01029 -0.00779 0.13563 -0.00610 -0.01191 0.20743 -0.00933 -1.43107 24.91423
0.00617 -0.00837 0.05052 0.00308 -0.00419 0.02526 0.00555 -0.00754 0.04547 0.00758 -0.01029 0.06207 0.00449 -0.00610 0.03681 0.00687 -0.00933 0.05630 0.82551 -1.12078 6.76170
0.54403 -0.00565 0.00326 1.10276 -0.01146 0.00661 0.69106 -0.00718 0.00414 0.74988 -0.00779 0.00449 1.23509 -0.01283 0.00740 0.92632 -0.00962 0.00555 0.97043 -0.01008 0.00582 52.50200
-0.00565 0.09840 -0.00443 -0.01146 0.19945 -0.00897 -0.00718 0.12499 -0.00562 -0.00779 0.13563 -0.00610 -0.01283 0.22338 -0.01005 -0.00962 0.16754 -0.00754 -0.01008 0.17552 -0.00790 -0.54543 9.49576
0.00326 -0.00443 0.02670 0.00661 -0.00897 0.05413 0.00414 -0.00562 0.03392 0.00449 -0.00610 0.03681 0.00740 -0.01005 0.06063 0.00555 -0.00754 0.04547 0.00582 -0.00790 0.04764 0.31463 -0.42717 2.57
714
0.83810 -0.00871 0.00502 0.66166 -0.00687 0.00397 1.01454 -0.01054 0.00608 1.08806 -0.01130 0.00652 0.88221 -0.00917 0.00529 1.24980 -0.01298 0.00749 1.32332 -0.01375 0.00793 1.10276 -0.01146 0.00
661 40.42794
-0.00871 0.15158 -0.00682 -0.00687 0.11967 -0.00538 -0.01054 0.18349 -0.00825 -0.01130 0.19679 -0.00885 -0.00917 0.15956 -0.00718 -0.01298 0.22604 -0.01017 -0.01375 0.23934 -0.01077 -0.01146 0.19945 -0.00
897 -0.42000 7.31199
0.00502 -0.00682 0.04114 0.00397 -0.00538 0.03248 0.00608 -0.00825 0.04980 0.00652 -0.00885 0.05341 0.00529 -0.00718 0.04330 0.00749 -0.01017 0.06135 0.00793 -0.01077 0.06496 0.00661 -0.00897 0.05
413 0.24227 -0.32893 1.98447
animal 0 AINV
at(Tr,1).leg(dim1,2).p2 2
27 0 US !GP
143.21391
-1.48783 25.90233
0.85825 -1.16523 7.02987
1.26450 -0.01314 0.00758 48.00642
-0.01314 0.22870 -0.01029 -0.49873 8.68266
0.00758 -0.01029 0.06207 0.28769 -0.39060 2.35647
1.42624 -0.01482 0.00855 1.29391 -0.01344 0.00775 42.58942
-0.01482 0.25796 -0.01160 -0.01344 0.23402 -0.01053 -0.44245 7.70292
0.00855 -0.01160 0.07001 0.00775 -0.01053 0.06351 0.25523 -0.34652 2.09057
0.70577 -0.00733 0.00423 0.58814 -0.00611 0.00352 0.69106 -0.00718 0.00414 168.04500
-0.00733 0.12765 -0.00574 -0.00611 0.10637 -0.00479 -0.00718 0.12499 -0.00562 -1.74579 30.39340
0.00423 -0.00574 0.03464 0.00352 -0.00479 0.02887 0.00414 -0.00562 0.03392 1.00705 -1.36727 8.24874
0.54403 -0.00565 0.00326 0.73518 -0.00764 0.00441 0.60284 -0.00626 0.00361 1.29391 -0.01344 0.00775 55.85395
-0.00565 0.09840 -0.00443 -0.00764 0.13297 -0.00598 -0.00626 0.10903 -0.00490 -0.01344 0.23402 -0.01053 -0.58026 10.10200
0.00326 -0.00443 0.02670 0.00441 -0.00598 0.03609 0.00361 -0.00490 0.02959 0.00775 -0.01053 0.06351 0.33472 -0.45445 2.74167
0.69106 -0.00718 0.00414 0.64695 -0.00672 0.00388 0.73518 -0.00764 0.00441 1.42624 -0.01482 0.00855 1.32332 -0.01375 0.00793 50.09990
-0.00718 0.12499 -0.00562 -0.00672 0.11701 -0.00526 -0.00764 0.13297 -0.00598 -0.01482 0.25796 -0.01160 -0.01375 0.23934 -0.01077 -0.52048 9.06130
0.00414 -0.00562 0.03392 0.00388 -0.00526 0.03176 0.00441 -0.00598 0.03609 0.00855 -0.01160 0.07001 0.00793 -0.01077 0.06496 0.30024 -0.40763 2.45923
0.52933 -0.00550 0.00317 0.39699 -0.00412 0.00238 0.49992 -0.00519 0.00300 0.61755 -0.00642 0.00370 0.48522 -0.00504 0.00291 0.63225 -0.00657 0.00379 175.45904
-0.00550 0.09574 -0.00431 -0.00412 0.07180 -0.00323 -0.00519 0.09042 -0.00407 -0.00642 0.11169 -0.00502 -0.00504 0.08776 -0.00395 -0.00657 0.11435 -0.00514 -1.82281 31.73434
0.00317 -0.00431 0.02598 0.00238 -0.00323 0.01949 0.00300 -0.00407 0.02454 0.00370 -0.00502 0.03031 0.00291 -0.00395 0.02382 0.00379 -0.00514 0.03103 1.05148 -1.42759 8.61267
0.44111 -0.00458 0.00264 0.57344 -0.00596 0.00344 0.48522 -0.00504 0.00291 0.57344 -0.00596 0.00344 0.70577 -0.00733 0.00423 0.64695 -0.00672 0.00388 1.29391 -0.01344 0.00775 60.73098
-0.00458 0.07978 -0.00359 -0.00596 0.10371 -0.00467 -0.00504 0.08776 -0.00395 -0.00596 0.10371 -0.00467 -0.00733 0.12765 -0.00574 -0.00672 0.11701 -0.00526 -0.01344 0.23402 -0.01053 -0.63092 10.98409
0.00264 -0.00359 0.02165 0.00344 -0.00467 0.02815 0.00291 -0.00395 0.02382 0.00344 -0.00467 0.02815 0.00423 -0.00574 0.03464 0.00388 -0.00526 0.03176 0.00775 -0.01053 0.06351 0.36395 -0.49413 2.
98107
0.52933 -0.00550 0.00317 0.48522 -0.00504 0.00291 0.57344 -0.00596 0.00344 0.66166 -0.00687 0.00397 0.58814 -0.00611 0.00352 0.72047 -0.00748 0.00432 1.42624 -0.01482 0.00855 1.33802 -0.01390 0.
00802 52.46081
-0.00550 0.09574 -0.00431 -0.00504 0.08776 -0.00395 -0.00596 0.10371 -0.00467 -0.00687 0.11967 -0.00538 -0.00611 0.10637 -0.00479 -0.00748 0.13031 -0.00586 -0.01482 0.25796 -0.01160 -0.01390 0.24200 -0.
01089 -0.54501 9.48830
0.00317 -0.00431 0.02598 0.00291 -0.00395 0.02382 0.00344 -0.00467 0.02815 0.00397 -0.00538 0.03248 0.00352 -0.00479 0.02887 0.00432 -0.00586 0.03537 0.00855 -0.01160 0.07001 0.00802 -0.01089 0.
06568 0.31438 -0.42684 2.57512
p2 0 ID

ASReml - RE: at(grp):member

Arthur — Thu, 26 Jan 2012 22:03:31 GMT

Dear Ian,

Var 200
...

at(Var)

works for me. There was originally a limit of 99, (probably in ASReml 2)
but that was reset to 999 some time back. I just tested it in ASReml 3 and ASReml 4 (aka 3.1).

Sorry for the delay in responding. Best wishes for 2012.

GenStat - RE: ordered categorical data with random effects - again

Peter Thomson — Wed, 18 Jan 2012 01:17:41 GMT

Hi Carole,
I'm not aware that you can fit this type of model within GenStat. However, this type of analysis is done very easily using the stand alone version of ASReml 3.0 (though not as yet included in ASReml-R)
Cheers, Peter

GenStat - ordered categorical data with random effects - again

Anonymous — Tue, 17 Jan 2012 02:43:51 GMT

Happy New Year

In October last year I posted a message seeking help with the procedure IRCLASS but got no responses.

I have another dataset with the same structure - ordered categories, treatments and replicates (fixed and random effects). The last data set was analysed using GLMM with a binomial distribution and considered the counts in each category (and cumulative counts). Out of curiosity I also used HGLM assuming there was no order to the categories.

Is there a better way to analysis this type of data? It seems like a standard data set so am interested in how other people analyse ordered categorical data with random effects.

Thanks again for any help
Carole

Post generated using Mail2Forum (http://www.mail2forum.com)

ASReml - at(grp):member

iwhite — Mon, 16 Jan 2012 15:07:30 GMT

Dear asreml experts,

I recently managed to fit a separate within-group variance for 500 sire families using formula diag(family):animal and renumbering animals within families (else run out of memory).

The task would have been more straightforward with formula at(family):animal but there seems to be a limit at 99 levels for the first factor. Is this a bug or a feature?

ASReml - RE: Model statement in multi trait analysis

Rodrigo — Fri, 13 Jan 2012 16:42:34 GMT

Thanks Dr. Arthur.
Regards
Rodrigo

ASReml - RE: Model statement in multi trait analysis

Arthur — Fri, 13 Jan 2012 09:40:41 GMT

Yes,
That includes a third trait (wt), adjusted for hy

ASReml - RE: Model statement in multi trait analysis

Rodrigo — Thu, 12 Jan 2012 16:32:04 GMT

Dear Dr. Arthur,

Thanks for you fast reply and valuable information... please just another question...
for add a new trait (wt) and new covariate (hy) to this model:

wgt lon ~ Trait at(Trait,1).age !r Trait.ANIMAL

i d write like this:

wgt lon wt~ Trait at(Trait,1).age at(Trait,3).hy !r Trait.ANIMAL is ok??

I d appreciatte any comments
Thanks in advance
Best Regards
Rodrigo

Forum | VSN International

ASReml - Singularity appearing in simple animal model

ASReml - RE: at(grp):member

ASReml - Modeling block-diagonal residual matrix in RR models

ASReml - RE: at(grp):member

GenStat - RE: ordered categorical data with random effects - again

GenStat - ordered categorical data with random effects - again

ASReml - at(grp):member

ASReml - RE: Model statement in multi trait analysis

ASReml - RE: Model statement in multi trait analysis

ASReml - RE: Model statement in multi trait analysis