[R] command Polygenic gives error message concerning dimensions of data

Veerle Bogaert Veerle.Bogaert at ugent.be
Wed Dec 24 14:52:33 CET 2008


Dear Sir/Madam,

Since a few day now I try to use the command "polygenic" from the GenAbel 
package. However, I keep bumping up against an error message: "Error in 
polygenic(Testo, kin = kinship, data = data1) : dimension of outcome and 
kinship.matrix do not match".

My data exists of 1240 individuals with 74 markers. It mainly consists of 
small families (2 or more brothers, father and mother; otherwise only the 
brothers are given). Genotypes are provided for the largest part of the 
members, whereas the phenotype is only given for the male individuals. I 
imported the data using the convert.snp.ped and load.gwaa.data commands. The 
kinship table was constructed with "makekinship". When I look at the 
dimensions of the kinship table, it is exactly the same as the number of 
individuals in my dataset.

I also provide you with the following information because maybe that gives a 
clue on where things are going wrong. When using the genetic control method 
via the qtscore command, I get the following message:" 
data1.qt<-qtscore(Testo~age+BMI,data1) Warning messages: 1: In qtscore(Testo 
~ age + BMI, data1) :  265 observations deleted due to missingness 2: In 
qtscore(Testo ~ age + BMI, data1) : Number of observations < 100, Lambda 
estimate is unreliable. Results for only a few SNPs are returned.

Can you tell me what I'm missing?

Thank you in advance,

Kindest Regards,


Veerle Bogaert
Ghent University Hospital
Veerle Bogaert
Ghent University Hospital
Department of Endocrinology 6K12 I.E.
De Pintelaan 185
9000 Ghent
Belgium
Tel +32 9 332 34 13
Fax +32 9 332 38 86



More information about the R-help mailing list