[vsni.co.uk] Contact us
Author Message
Post new topic

  ASReml  ~  Format of pedigree data in BLUP-analysis

Christof Pietsch
Posted: Tue Oct 13, 2009 8:40 am Reply with quote
Joined: 15 Jun 2009 Posts: 8
Dear all,
I'd like to know, if entries with no pedigree information but with phenotypes have to be incorporated in the pedigree-data for the calculation of the numerator relationship matrix in a BLUP-analysis using the VPEDIGREE-directive.
Thank you very much in advance!
Christof
View user's profile Send private message
Arthur
Posted: Tue Oct 13, 2009 10:57 pm Reply with quote
Joined: 05 Aug 2008 Posts: 471 Location: Orange, NSW
Dear Christof,

YES.
The NRM matrix must incorporate all the individuals with data
as well as the known parents of those individuals.

The NRM is calculated independent of the data, and then associated with it. The association fails if the data contains individuals not in the NRM.

_________________
Arthur Gilmour

Retired Principal Research Scientist (Biometrics)
View user's profile Send private message Send e-mail Visit poster's website
gdutkowski
Posted: Wed Oct 14, 2009 12:13 am Reply with quote
Joined: 11 Aug 2008 Posts: 15 Location: Hobart, Australia
Joao Costa e Silva has found in the forestry context that including such individuals in the analysis can lead to quite large differences in the estimates of the additive variance. We standardly apply an extra units variance with a possible positive or negative variance (sorry, don’t know the R way of doing that) to such analysis to avoid bias in the variance estimates. Or we exclude them from variance estimation.
I don’t know why this happens, but presumably as individuals with no parents these are treated as unrelated founders in the pedigree and are somehow given high weight in the variance estimation. They may, in fact, be related, or they may come from different populations.

Greg Dutkowski
Quantitative Geneticist
PlantPlan Genetics http://www.plantplan.com
c/- School of Plant Science
University of Tasmania
Private Bag 55
HOBART 7001

+61-3-62262213 (w) +61-408-238495 (mob) +61-3-62262698 (fax)
From: ASReml users discussion group [mailto:ASREML-L@DPI.NSW.GOV.AU] On Behalf Of Christof Pietsch
Sent: Tuesday, 13 October 2009 7:41 PM
To: ASREML-L@DPI.NSW.GOV.AU
Subject: Format of pedigree data in BLUP-analysis



Dear all,
I'd like to know, if entries with no pedigree information but with phenotypes have to be incorporated in the pedigree-data for the calculation of the numerator relationship matrix in a BLUP-analysis using the VPEDIGREE-directive.
Thank you very much in advance!
Christof









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.





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.

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

_________________
Greg Dutkowski
PlantPlan Genetics
c/- School of Plant Science
University of Tasmania
Locked Bag 55
HOBART 7001
+61-3-62262213 (w) +61-408-238495 (mob) +61-3-62262698 (fax)
View user's profile Send private message Send e-mail Visit poster's website
Bruce Southey
Posted: Wed Oct 14, 2009 4:52 pm Reply with quote
Guest
Hi,
Basically, as Greg said, individuals without a pedigree are treated as founders yet they are not founders. These individuals should appear as outlier (compared to the real founders) because they potentially include all the genetic changes since the population was founded. Since the variance component estimates only apply to the base/founder population, these individuals artificially increase estimates because that genetic change is unaccounted for. Thus the requirement of genetic groups.

For example, see
C. Pieramati and L. D. Van Vleck 1993. Effect of genetic groups on estimates of additive genetic variance
http://jas.fass.org/cgi/content/abstract/71/1/66

Díaz, C.; Moreno, A.; Carabaño, M. J. 2002. Effects of genetic groups on estimates of variance components for a maternally influenced trait in the Spanish Limousine
Journal of Animal Breeding and Genetics, Volume 119, Number 5, October 2002 , pp. 325-334(10)
http://www3.interscience.wiley.com/journal/118922871/abstract?CRETRY=1&SRETRY=0

If you have very few unknown individuals relative to known contemporary individuals then you should just remove these unknown individuals as the contemporary mean should be well estimated.

Bruce



On 10/13/2009 06:19 PM, Greg Dutkowski wrote:
Quote:
<![endif]--> <![endif]-->
Joao Costa e Silva has found in the forestry context that including such individuals in the analysis can lead to quite large differences in the estimates of the additive variance. We standardly apply an extra units variance with a possible positive or negative variance (sorry, don’t know the R way of doing that) to such analysis to avoid bias in the variance estimates. Or we exclude them from variance estimation.
I don’t know why this happens, but presumably as individuals with no parents these are treated as unrelated founders in the pedigree and are somehow given high weight in the variance estimation. They may, in fact, be related, or they may come from different populations.

Greg Dutkowski
Quantitative Geneticist
PlantPlan Genetics http://www.plantplan.com
c/- School of Plant Science
University of Tasmania
Private Bag 55
HOBART 7001

+61-3-62262213 (w) +61-408-238495 (mob) +61-3-62262698 (fax)
From: ASReml users discussion group [mailto:ASREML-L@DPI.NSW.GOV.AU (ASREML-L@DPI.NSW.GOV.AU)] On Behalf Of Christof Pietsch
Sent: Tuesday, 13 October 2009 7:41 PM
To: ASREML-L@DPI.NSW.GOV.AU (ASREML-L@DPI.NSW.GOV.AU)
Subject: Format of pedigree data in BLUP-analysis



Dear all,
I'd like to know, if entries with no pedigree information but with phenotypes have to be incorporated in the pedigree-data for the calculation of the numerator relationship matrix in a BLUP-analysis using the VPEDIGREE-directive.
Thank you very much in advance!
Christof









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.





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.





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.

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

Display posts from previous:  

All times are GMT
Page 1 of 1
Post new topic

Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You can attach files in this forum
You can download files in this forum