I want to use global test.
1
0
Entering edit mode
Kim • 0
@fa33c3d0
Last seen 23 months ago
Cheongju-Si

Hello, I want to get the p-value of each probe through a global test with the data in https://github.com/Foreist/dummy-data .

The data source for the above address is https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE13351.

I selected a specific probe and normalized it with python.

I inevitably have to use rpy, but even if it doesn't, I'm curious how to apply that data to the global testing technique even through r.

I mostly use Python, so I'm not familiar with r.

There is really no place to ask.

You can also recommend an example.

gene r clst rpy globaltest • 610 views
0
Entering edit mode
@gordon-smyth
Last seen 15 minutes ago
WEHI, Melbourne, Australia

Dear Kim,

I understand that you're just getting started with expresson analyses, but I think that you're on the wrong track in several ways:

1. globaltest (the Bioconductor package) does not give p-values for individual probes. It is only for gene sets.

2. You can't normalize an individual probe in isolation. All the probes need to normalized together. The GSE data series you link provides normalized expression values, so there is no need to you to re-normalize the values.

3. This forum is only for Bioconductor R software. Bioconductor provides functionality that is not available in any Python package.

4. There are many Bioconductor guides describing how to analyze Affymetrix data. Probably the easiest would be to download the normalized expression values from GEO and plug them into a limma package analysis.

0
Entering edit mode

The data that I use is notification is GPL96.

Should I run it like gtKEGG (set, plz_gt, id = "pd.hg.u133.plus.2")?

alias     direction     p-value   Statistic Expected


M92287_at CCND3 assoc. with response = ALL 2.06e-07 53.19830 2.7

D38073_at MCM3 assoc. with response = ALL 2.48e-06 46.43408 2.7

M81933_at CDC25A assoc. with response = AML 2.18e-05 39.79937 2.7

U31814_at HDAC2 assoc. with response = ALL 3.58e-05 38.17296 2.7

L41870_at RB1 assoc. with response = ALL 3.82e-05 37.95700 2.7

Also, I know that the results are shown in the picture above, but is it correct that you do not provide p-value for individual probes?