survival analysis of TCGA data

Question

[monukmr98]

Hi there

I am doing some survival analysis on TCGA patients of my interest. I have to draw Kaplan-meier curve, which I am able to do with "survival" package in R using their default "aml" data. Same format I applied on my samples' data and got some curve. But, I have doubt whether I am doing it right way or not.

e.g. survival package has default data format as follows

> aml
   time   status          x
1     9      1    Maintained
2    13      1    Maintained
3    13      0    Maintained
......

and my prepared data format for GBM tissue samples is as follows

     time         status      x
1    11.9333      1    class1
2    4.8000      1    class2
3    18.6000      1    class1
....

I have doubt regarding status tag (middle column i.e. all are 1 ) in my data; as my data does not contain information of censored subjects. I have taken all dead patients and time is reported in months for corresponding dead patients (days_to_death column). All this data has been taken from clinical_patient_gbm.txt file. So my question is "Am I MISSING SOMETHING" ? like "Censored data". If yes, then how to obtain censored data from TCGA and from which file (followup file)?

Thanks

[russhh]

Why have you biased the dataset to the dead individuals only? Have a look for days_to_last_follow_up or similar to complement your current dataset

[Miao Yu]

You mean he should add the days_to_last_follow_up to the current dataset as censored data(status is '0')?