[BioC] best algorithm Affy low end signal

[Anthony Bosco]

Hi all.

1) I would like to ask for advice on the best algorithm to study 
genes at the low signal end on Affy chips.

I have tried

rma
gcrma
vsn (vsn norm, nbg, medianpolish)
liwong
mas5

To be honest, although mas5 is very noisy at low end, the real genes 
(the ones that I know are important and should be there) where 
clearly out of the noise, where as  some of the other methods seem to 
bury some (but not all) of these genes (in poarticular gcrma seems to 
bury some genes in the noise at the low end).


2) Reporting fold change

The fold change values are dependant on the analysis algorithm, in 
mas5 the fold changes are very high at low end, but not in rma.

Is this real (ie I would expect in any assay that the low end signal 
is more noisy and variable at low end,  therefore low end genes would 
have to have large fold change to stand out of noise on M vs A plot)


Any advice gratefully recieved


Anthony
-- 
______________________________________________

Anthony Bosco - Cell Biology Research Assistant

Institute for Child Health Research
(Company Limited by Guarantee ACN 009 278 755)
Subiaco, Western Australia, 6008

Ph 61 8 9489  , Fax 61 8 9489 7700
email anthonyb at ichr.uwa.edu.au