This archive contains the level 3 data of the 215 gliolastoma multiforme (GBM) 
samples of the TCGA project and 51 
control samples 
normally used by the data generating center - the RPPA Core Lab
 at the University of Texas 
MD Anderson Cancer Center, Houston, TX.



The data were loading corrected, meaning different samples have the 

same average total amount of proteins after this correction.
The method is described as follows:



1)starting with the log2 raw protein concentration data, median-center the data of each slide;  

2)from 1), calculate the median value for each sample;

3)subtract the medians in 2) from the raw concentration data of the corresponding sample.




Users are free to try additional methods starting from the Level 2 data, such as median 
centering each sample, 
median polish, etc.
 



The samples whose names start with "Mixed-Lysate", "MDA-MD", "Jurkat" are 
positive control samples. 
The mixed lysates are a pool of multiple cell lines. 
These samples could be used for batch effect correction or 
help evaluate data quality.


One sample (UUID=0addd9ee-dd84-44d7-baa9-6ceb1970a97f) is not recognized by the Biospecimen Metadata Browser, and is not included here, but it will be
included in a revised archive when the cause is found out.
