Regarding Select Genomic Profiles

[Peixin Dong]

Dear cBioPortal group,

 

I am exploring the expression of the TTK gene and TIMP3 gene in Uterine Corpus Endometrial Carcinoma (TCGA, PanCancer Atlas), trying to find:

 

1.     whether those endometrial patients with high TTK expression would express less TIMP3.

2.     whether TTK is overexpressed, while TIMP3 is downregulated in endometrial cancer tissues relative to normal tissues

 

Could you kindly please let me know the difference between:

 

1.     mRNA expression z-scores relative to normal samples

2.     mRNA expression z-scores relative to all samples

 

In my case, which one I should select?

 

Thank you so much for your help.

 

Peixin Dong, MD, PhD 

Department of Obstetrics and Gynecology 

Hokkaido University School of Medicine 

Hokkaido University 

N15, W7, Kita-ku, Sapporo, 0608638, Japan 

Tel: +81-11-706-5941; Fax: +81-11-706-7711 

E-mail: dpx...@med.hokudai.ac.jp; E-mail: dpx...@gmail.com

 

[Benjamin Gross]

Hi Peixin,

mRNA expression z-scores relative to all samples are computed expression z-scores (log-transformed) for gene/sample compared against the expression distribution of _all_ samples in the cohort.

mRNA expression z-scores relative to normal samples are computed z-scores for gene/sample compared against expression of  _all_adjacent _normal_ samples in the cohort.

I’m not a biologist, generally speaking I think comparing expression against adjacent normal tissue is a better control, gives a clearer signal.  Someone else on the team may have more to say.

I hope this helps,

-Benjamin

-- 
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.
To view this discussion on the web visit https://groups.google.com/d/msgid/cbioportal/000001d8b28e%2410ae2ce0%24320a86a0%24%40med.hokudai.ac.jp.