dChip does not take all files

[Arshi]

Hi,

I am using data from ebi. id egeod-10601. It has 61 arrays in it and
dChip only takes 18 samples in it, and return CEL images of these 18
images. Why is it doing so ?

Also, to normalize and data or to perform any analysis is it important
to do the make information file, where one puts the process.ontology
files. Because when I try to do that, dChip crashes. I am using the
2008 version.

Also, I am working with time series data, and the CEL image of such
data is streaked with yellow lines, and not just yellow dots. Does
that mean that the data i snot good ?

It will be great if I can get these answers

Thanks

Arshi

[Cheng Li]

Arshi,

Most likely the arrays you are using are not the same type. For U95 and V2
data, see
http://biosun1.harvard.edu/complab/dchip/combine%20chip.htm#u95

No need to make info file to normalize. Do not use GO files or use older
files.

Yellow lines are fine since probes of the same probe set on older arrays are
not scrambled.

Cheng

[Arshi]

Hey,

Thank you Cheng.

I am able to run all the files. The samples which I was taking had
both antisense and 2.0 affymetrix arrays.

Thanks !

Arshi

On Oct 20, 8:58 pm, "Cheng Li" <lch3...@gmail.com> wrote:
> Arshi,
>
> Most likely the arrays you are using are not the same type. For U95 and V2

> data, seehttp://biosun1.harvard.edu/complab/dchip/combine%20chip.htm#u95