Dear Alistaire,
1
The primary reason for the analysis ignoring the pedigree is that
the pedigree applies to TreeID but
TreeID is not fitted in the model.
So, you will need to include TreeID in the model instead of Family.
Furthermore, I would use the second Pedigree file as I think the
genetic group effects will be easier to interpret.
2
I gather you have a fully nested structure of
Trees within Families within groups.
Consequently, a equivalent model would be be achieved by
creating a data column for GROUP and fitting
Weight ~ mu c(GROUP) !r Block Family
Also, I suspect Block could reasobnably fitted as a random effect.
3
Alternatively, you could just define the pedigree for Family
and put the !P on the Family line rather than the TreeID line
Best wishes.
Arthur
> Hi,
>
> I have been trying to fit genetic groups, without luck.
>
> I have data from an apple tree breeding population consisting of open
> pollinated families. The mother trees were divided into genetic group based
> on species, region and era of development. There are 36 genetic groups.
> Because the mother trees were planted in tree repositories surrounded by
> unkown trees, potentially from any of the 36 genetic groups, I have assumed
> that each family has 2-3 unknown pollen donors. I have tried the analysis
> with pollen donors set to 0 (unknown genetic group) and with pollen donors
> set to genetic group37.
>
> An example of the program file:
>
> Title: Univariate variance component estimation - Apple data
> TreeID !P
> Site 2 !A
> Row 40
> Tree 400
> Family 225 !A
> Block 257
> Origin 100 !A
> Weight
> pedigree.txt !ALPHA !MAKE !GROUPS 36 !NEW
> data.txt !filter 2 !select 1 !skip 1 !maxit 10
> Weight ~ mu Block !r Family
>
> I have tried two pedigree files. The first has zeros for the pollen donor
> as it could belong to any of the 36 genetic groups:
>
> group1 0 0
> group2 0 0
> ...(etc.)
> group36 0 0
> fam1 0 group1
> fam2 0 group1
> ...(etc.)
> fam225 0 group36
> ID1 0 fam1
> ID2 0 fam1
> ...(etc.)
> ID4000 0 fam5
>
> The second pedigree file has group37 assigned to the pollen donors.
>
> group1 0 0
> ...(etc.)
> group37 0 0
> fam1 group37 group1
> fam2 group37 group1
> ...(etc.)
> fam225 group37 group36
> ID1 group37 fam1
> ID2 group37 fam1
> ...(etc.)
> ID4000 group37 fam5
>
> However, the answers I get are the same as if I run the program without the
> pedigree file and !P option. The first question is how do I fit genetic
> groups ? The second question is can I fit genetic groups to only the dams,
> leaving the pollen donors unknown, or do I have to fit a genetic group to
> the unknown pollen donors?
>
> Thanks in advance for any suggestions,
> Alastair Currie
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
Arthur Gilmour PhD email: Arthur.Gilmour@agric.nsw.gov.au
Senior Research Scientist (Biometrics) fax: <61> 2 6391 3899
NSW Agriculture <61> 2 6391 3922
Orange Agricultural Institute telephone work: <61> 2 6391 3815
Forest Rd, ORANGE, 2800, AUSTRALIA home: <61> 2 6362 0046
ASREML is currently free by anonymous ftp from pub/aar on ftp.res.bbsrc.ac.uk
or point your web browser at ftp://ftp.res.bbsrc.ac.uk/pub/aar/
To join the asreml discussion list, send the message
subscribe
to asreml-request@chiswick.anprod.CSIRO.au
The address for messages to the list is asreml@chiswick.anprod.CSIRO.au
<> <> <> <> <> <> <>
"Ye are not your own for ye are bought with a price: therefore glorify God
in your body, and in your spirit, which are God's" 1 Corinthians 6:20
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>