About retrieve mutation information: "the barecode can be used to match images to sample."

ncoudray / DeepPATH

Classification of Lung cancer slide images using deep-learning

492 stars 213 forks source link

About retrieve mutation information: "the barecode can be used to match images to sample." #88

Closed agentdr1 closed 3 years ago

agentdr1 commented 3 years ago

Hi,

I am working on a similar mutation prediction task (TCGA colon) and closely follow your readme (3.1 - Retrieve mutation information). I have the WSIs, the related manifest file and also the MAF file as described in 3.1; you wrote: "and the barecode can be used to match images to sample. " Which barcode, or in other words IDs do you use for the slides and which IDs from the MAF file to match them?

Best regards, Daniel

ncoudray commented 3 years ago

Hi Daniel,

The first 12 digits of the image name (TCGA-##-####); you can check https://docs.gdc.cancer.gov/Encyclopedia/pages/TCGA_Barcode/ for barecode explanations.

Best, Nicolas

agentdr1 commented 3 years ago

That was exactly what I need, thanks for the quick reply!