normalized intensities

[Cheng Li]

-- Forwarded old discussion --

Posted: 28 Dec 2007 11:05 pm Post subject: normalized intensities

--------------------------------------------------------------------------------
Hi, Cheng,

How can we output the normalized intensities through dChip? I mean the
intensities just after the invariant set normalization before we fit
the model.

Thank you very much,

Maymay


Posted: 29 Dec 2007 01:04 am Post subject:

--------------------------------------------------------------------------------
you can use "Image/export cel":
http://www.dchip.org/image%20view.htm#export_cel

cheng


Posted: 01 Jan 2008 03:11 pm Post subject:

--------------------------------------------------------------------------------
Thank you very much for your reply, Cheng. Also can you tell me how I
can get the txt form from CEL files. Since I need to do some analysis
in R, I need to import the probe intensities into R.

Thank you,

Maymay


Posted: 02 Jan 2008 05:15 pm Post subject:

--------------------------------------------------------------------------------
dchip exported CEL files are in text format. If affy CEL files are in
binary format, you can use the conversion tool:
http://www.dchip.org/prepare%20data.htm

cheng


Posted: 03 Jan 2008 04:19 pm Post subject:

--------------------------------------------------------------------------------
Thank you, Cheng. I got those files. They are .cel file, right?
But how can I read them into R? They don't seem like tab-delimited
files. Incorrect dimensions

Beibei


Posted: 03 Jan 2008 04:25 pm Post subject:

--------------------------------------------------------------------------------
They are text files and you can open in text editor to check their
format. Alternatively, you may try "Tools/export probe set" to export
probe data for analysis in R.

cheng


Posted: 30 Jan 2008 03:44 pm Post subject:

--------------------------------------------------------------------------------
Hi Beibei,

you can read those using the aroma.affymetrix R package (http://
www.braju.com/R/aroma.affymetrix/). You find information on how to
import dChip-exported CEL files at:

http://groups.google.com/group/aroma-affymetrix/web/import-from-dchip

Note that some chip types are rotated by dChip and remains rotated
when exported. The above package tries to detect this and rotate data
back while "importing" the data. Importing here means reading ASCII
CEL files and storing them as binary CEL files (which are faster to
read).

The structure of data in CEL files per se are quite straightforward,
but when coupled with the CDF files, which defines what probes go with
what probesets, it becomes much more tricky, e.g. the data is no
longer in nice tabular format but quite nested. So be prepared for
that if you want to do your own low-level access of data.

Please address further aroma.affymetrix questions to the dedicated
mailing list.

Question/suggestion to Cheng Li: Would it be hard to rotate those CEL
files back before exporting or add a field to the CEL header
specifying the amount of rotation done?

Hope this helps,

/Henrik