Re: Asreml for phenotypic plasticity

From: Melissa Gunn <M.Gunn_at_SHEFFIELD.AC.UK>
Date: Wed, 29 Mar 2006 14:40:42 +0100

Hi all,
I am trying to analyse variation in phenotypic plasticity using Asreml. I have
multiple measurements of parturition date for ~270 females, and I am using the
interaction between food abundance (CONES) and individual identity as the slope
of the reaction norm (i.e. CONES.newid). I find I get quite different answers
in the asr file if I used standardised cones (centred on zero). Should this be
the case?

Also, I find I have only one estimated effect size for CONES in the sln file,
although AGE (also a fixed effect) has an effect size for each age presented.
Would making CONES a factor change this?

I would also like to know how the analysis of variance presented in the asr file
relates to Wald stats (if it does!). Can I use this to compare models with
different fixed effects?

Finally, I find no significant variation in phenotypic plasticity - but when I
plot the values from the yht file for the best-fitting model, it looks as
though there is significant variation. Can anyone suggest why this would be?

My job file is as follows:
 newid !P #identity of individual
 YR !A #year of parturition
 AGE !A #age of individual
 CONES #measure of food abundance
 PD #parturition date in Julian days
 DAM !P #identity of individual's mother
 sCONES #standardised cone number
 sAGE #standardised age of individual
 Byear !A #birthyear of individual
PD ~ mu AGE sCONES !r newid DAM CONES.newid YR Byear
1 1 5 !STEP 0.001
0 0 IDEN 382

newid 1
newid 0 IDV 56

DAM 0 IDV 21

CONES.newid 2

YR 1
YR 0 IDV 21

Byear 1
Byear 0 IDV 21 #!GP

Thanks in advance for any suggestions.

Dr. Melissa Gunn
Department of Animal and Plant Science
University of Sheffield
Western Bank
Sheffield S10 2TN
United Kingdom
Ph. +44(0)114 2220106
Fax: +44(0)114 222 0002
Received on Mon Mar 29 2006 - 14:40:42 EST

This webpage is part of the ASReml-l discussion list archives 2004-2010. More information on ASReml can be found at the VSN website. This discussion list is now deprecated - please use the VSN forum for discussion on ASReml. (These online archives were generated using the hypermail package.)