Redo normalization and modeling

[Cheng Li]

Sam,

> the modeling and normalization takes place across all samples. Is
this correct?

This is correct. But as long as you have removed outlier samples after
the first run of modeling, and redo normalization and modeling, you
don't have to renormalize. You can always open all the good samples,
and use "array list file" to specify what samples to analyze and
"sample info file" (Ploidy column) to specify normal samples.

When you add new samples, you need to redo normalization and modeling.

Cheng


-----Original Message-----
From: Sam Volchenboum
Sent: Wednesday, December 16, 2009 4:58 PM
To: lch...@gmail.com
Subject: Quick question

Cheng,

Regarding the dcp files that dChip creates:
Do I need to re-run the normalization and MBEI whenever I want to add
or subtract samples from the analysis?

In other words, I am trying out different ways of analyzing copy
numbers, both including normal HapMap controls and not. So, my
impression is that I need to run the whole "normalization and MBEI"
with all the samples in the working folder, each time I want to add or
remove samples, since the modeling and normalization takes place
across all samples. Is this correct?

Thanks for your advice.

-sam

[Renzoni, Elizabeth]

Dear Cheng,

I wonder if I could ask your help for what is probably a very simple issue. Across chips, I would like to find all the genes that have a similar expression pattern to one gene selected within a clustering page. The online manual suggests going to "Clustering", and then clicking on "similar profile", but under the clustering tab I cannot find this command, nor can I find it elsewhere. Is there a something very basic that I am missing?

Thank you for your help

Elizabeth

Sam,

Cheng

Cheng,

Thanks for your advice.

-sam

--
This message is from the dChip Software user group.
-- If you ask dChip questions, please include dChip output messages, generated files, or screenshots to help rapid debugging.
-- To post to this group, send email to dchip-s...@googlegroups.com
-- To unsubscribe from this group, send email to dchip-softwar...@googlegroups.com
-- For the latest updates or to change email options, visit this group at http://groups.google.com/group/dchip-software