Dear List,
I am trying to run a mixed model which, on the R console, prints output as
follows:
[1] "Marker"
[1] "perm no."
[1] NA
Linear mixed model fit by REML
Formula: peg.no.prm ~ 1 + (1 | family/f)
Data: modeldf
AIC BIC logLik deviance REMLdev
3119 3134 -1555 3112 3111
Random effects:
Groups Name Variance Std.Dev.
f:family (Intercept) 0.0 0.000
family (Intercept) 0.0 0.000
Residual 178.5 13.360
Number of obs: 388, groups: f:family, 73; family, 60
Here 'Marker' is one of 426 markers, and 'perm.no' is one of
10,000
permutations of variance for each marker.
Each time R begins with a new marker column, it prints 'Marker', and for
each permutation it prints a number from 1 to 10000 (it hasn't here, but
even 'NA' tells me where one data block begins and another ends-- MUCH
easier for eyeballing).
Of course, since at the end, I will have 426*10000 blocks of data, the
console doesn't handle it and I tried printing out to a text file (.txt)
using capture.output().
The problem with this is that in the text file, my labels ('Marker' and
'Perm no.') aren't printed (since the capture command only uses
output from
summary(*model*).
This leaves me with about 4 million blocks of output (hypothetically) and
with no way of telling where one marker ends and the next begins.
Eyeballing this will take years.
Is there a way to therefore produce labeled output while writing to a text
file?
Thanks a lot for any suggestions that might help,
Aditi
---------------------------------------------------------
Code for a test file with 10 markers:
model<-read.table(".......)
modeldf<-data.frame(model)
modeldf[2:13]<-lapply(modeldf[2:13],factor)
colms<-(modeldf)[4:13] ## ten marker columns
se<-c(1:10000)
peg.no<-(modeldf)[,14]
library(lme4)
for(f in colms)
{
print("Marker")
{
for( i in 1:10000)
{
print("perm no.")
print(se[i])
{
peg.no.prm<-sample(peg.no, length(peg.no))
try(fit5<-lmer(data=modeldf, peg.no.prm~1 + (1|family/f)))
print(summary(fit5))
capture.output(fit5, file="testperm5.txt", append=T)
}}}
}
The data files are at:
<http://www.4shared.com/file/131980362/460bdafe/Testvcomp10.ht
ml> (excel)
http://www.4shared.com/file/131980512/dc7308b/Testvcomp10.html
(txt)
----------------------
A Singh
Aditi.Singh at bristol.ac.uk
School of Biological Sciences
University of Bristol
