Question about Differential Gene Expression Analysis with GEO2R

Question

[vin.darb]

Hello !

For a project I reanalyse data from a study (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE50395) in order to perform a Differential Gene Expression Analysis between 2 conditions (3 patients vs 3 controls).

My first idea was to use the software GEO2R because data values are already normalized signal intensity from the single-ship array.

But as you can see with the value distribution box-plots:

https://www.ncbi.nlm.nih.gov/geo/geo2r/backend/geo2r.cgi?ctg_time=1543321888&job_key=JSID_01_637786_130.14.22.10_9000_geo2r

Values across samples aren't median-centered ..

My question is does GEO2R standardize theses values between the different samples or I have to normalize the data first? If I have to normalize the data, how to perform this normalization?

Thank you in advance for your help