Re: nested genotype effects

Dear Craig,

If you have Genotypes nested in (classified by) Families and
fit both Family and Genotype as random, the BLUPs for genotypes
within any particular family will not in general sum to zero.

You will find that if the Family effect is positive, the average BLUP
for genotypes within the familiy is also positive.

You may get BLUPS averaging zero within family if Family is fitted
FIXED.

The result that BLUPs sum to ZERO only applies in simple (uncorrelated
cases). It does not apply to an 'individual' model.

Given the nested structure of Family and Genotype, there is no point
forming Family.Genotype interaction. It is exactly the same Design
matrix as 'Genotype' except for a huge number of ZERO columns being
inserted.

Recoding the data will not change the result of BLUPs not summing to
ZERO.

I trust this helps.

On Thu, 2009-03-26 at 11:32 +1000, Craig Hardner wrote:
> Hi Folks
>
>
>
> I am trying to run an unbalanced MET model where I have genotypes
> nested within families. (This is a hybrid design with not many parents
> so I don’t want to implement a individual model). I have 892 genotypes
> in total across 10 families. I want to get my blups for genotype
> within families centred around 0 in each family
>
> I tried Trial.Fam+Trial.Genotype bit this doesn’t centre my genotype
> within family blups around 0.
>
> If I try a model with Trial.Fam+Trial.Fam.Genotype with
>
> Trial.Fam.Genotype 2
> Trial 0 Diag !+22
> 22*1
> Fam.Genotype 0 ID
>
> I get a 196240 effcets (22 trials*10 families *892 genotypes) for the
> Trial.Fam.Genotype (instead of 19624, 22*892) and the model doesn’t
> converge.
>
> I imagine I could go and code genotype 1..n for each family but is
> there another way to reduce the number of effects so hopefully I get
> convergence?
>
> I tried a similar model in ASreml-R and got Fault 3125 Failed forming
> G scores
>
> Thanks for your help
>
>
>
>
>
> Dr Craig Hardner
>
> Research Fellow
>
> School of Land, Crop and Food Sciences
>
> University of Queensland
>
> St Lucia 4067 Queensland, Australia
>
> ph: +61 7 3346 9465
>
> email: craig.hardner_at_uq.edu.au
>
>
>
>
>
>
>
>
> This message is intended for the addressee named and may contain confidential information. If you are not the intended recipient, please delete it and notify the sender. Views expressed in this message are those of the individual sender, and are not necessarily the views of their organisation.

-- 
May the God and Father of the Lord Jesus Christ guide and bless you.
Arthur Gilmour
Mobile Number +61 427 227 468
Home phone +61 2 6364 3288  Skype: Arthur.Gilmour
http://www.CargoVale.com.au
Wisdom is justified by all her children  Luke 7:35
Travel:
Katoomba 13-15 March
USA  1-11 April
Brazil/UK July/Aug ??