Re: Information about TCGA Agilent Gene Expression Data, in cBioPOrtal

[Nikolaus Schultz]

Dear Luca,

All Agilent microarrays in TCGA are tumor vs pooled normal (some kind of pooled Promega reference RNA). 

Niki. 

On Oct 11, 2013, at 8:50, Luca Zammataro <luca.za...@ieo.eu> wrote:

Good afternoon,

My name is Luca, I'm bioinformatician at Italian Institute of Technology, IIT in Milan, Italy.

I'm writing you, because I wish to understand in what way cBioPortal calculates the mRNA expression levels for "X genes" using integrated values from the Affymetrix and the Agilent technologies (platforms). 

Affymetrix apart, in which you can work on absolute values, the problem rises when you want obtain absolute values also for the Agilent.

In fact, studying the TCGA expression data produced by means of the Agilent Technology, I'm trying to understand how the log2 lowess normalized (cy5/cy3) values were produced: 

In few words, I wish to understand if the cy5 value correspond to the analyte from the healthy portion of the sample tissue, whereas the cy3 correspond to the analyte from the tumor of the same sample. (i.d. Ovarian Cancer)

Or if the healthy sample was taken for a normal tissue that the TCGA consider as a "referent  normal sample", and this is the same for all the tumor sample within the TCGA Ovarian Cancer sample list.

The question is inherent to the possibility to normalize all the log2 ratio from all the OV samples..

Please, Can you help me to understand?

Should I consider the mRNA expression values, provided by the  cBioPortal such as a normalization of values, taking in consideration the cy3/cy5 ratio, such a ratio between an Unique Normal Sample and cancer expression values, for all the cancer samples in TCGA?

Sorry for the disturb,

Best regards.

Luca

Center of Genomic Science of IIT@SEMM

Luca Zammataro, MD, PhD
Computational Research Unit
Luca.Za...@iit.it

site: 

http://www.iit.it/en/people/iit-semm/post-doc/luca-zammataro.html

[sasha...@gmail.com]

Hello Niki,

I've been trying to match gene expression file (agilent) to their proper tumor and normal draw. So, if it is pooled - Do you know which normal used for pool (blood, tissue?) ? And maybe you have more specifics data on what is that normal pool is?

Thank you,
Sasha
PhD Student
BGU, Israel

[Nikolaus Schultz]

It is Stratagene Human Reference RNA

http://www.genomics.agilent.com/article.jsp?pageId=1452

--
You received this message because you are subscribed to the Google Groups "cBioPortal for Cancer Genomics Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cbioportal+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Sasha Kugel]

Sorry to bug you again, but maybe you could help me a bit more:

Does that mean the normal samples were also referenced to the Stratagene Human Reference RNA?

Thank you again in advance,

Sasha

On Tue, Sep 23, 2014 at 7:59 AM, Sasha Kugel <sasha...@gmail.com> wrote:

Thank you very much!

that explains the "Stratagene Univeral Reference" in the file descriptions.

Have a good day,
Sasha

--
You received this message because you are subscribed to a topic in the Google Groups "cBioPortal for Cancer Genomics Discussion Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cbioportal/ROcju67fX_4/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cbioportal+...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Aleksandra (Sasha) Kugel
The Shraga Segal Dept. of Microbiology, Immunology and Genetics
The Faculty of Medical Sciences
Ben Gurion University of the Negev

--

Aleksandra (Sasha) Kugel
The Shraga Segal Dept. of Microbiology, Immunology and Genetics
The Faculty of Medical Sciences
Ben Gurion University of the Negev

[Sasha Kugel]

Thank you very much!

that explains the "Stratagene Univeral Reference" in the file descriptions.

Have a good day,
Sasha

On Tue, Sep 23, 2014 at 12:49 AM, Nikolaus Schultz <nsch...@gmail.com> wrote:

--
You received this message because you are subscribed to a topic in the Google Groups "cBioPortal for Cancer Genomics Discussion Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cbioportal/ROcju67fX_4/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cbioportal+...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Nikolaus Schultz]

Yes. All samples run on the Agilent expression arrays were co-hybridized with and normalized to the same universal reference sample. 

Niki.