Hello bioinformaticians,
I've recently downloaded few RNAseq data files from Genomic Data Commons Data (GDC) portal. These files belong to Broad institutes CCLE project.

Now the problem is that GDC portal do not provide CCLE cell line names:

enter image description here

So my question is how could I link the GDC ids (prefixes of File Names) to CCLE cell line names?

I would imagine because it is an important step in the analysis, there should be some way to do that.

  • $\begingroup$ Where have you searched? Is there any information page about how to identify the cell lines or the files? $\endgroup$ – llrs Dec 11 '18 at 8:52
  • $\begingroup$ It's an issue with GDC portal. So I looked around on their website. There is not even an email address to query about this issue. $\endgroup$ – rraadd88 Dec 11 '18 at 11:30

The filenames of those bams contain the cell line names. They are constructed:


Thus the first 5 cell lines in your screen capture are SK-N-F1, TE-9, CJM, HCI-H1915 and CAL-33.

All the information that is need is also present in the experimental metadata xml file attached to each sample record. This is accessible by downloading it from the record on the website, or by querying the GDC metadata API (https://gdc.cancer.gov/developers/gdc-application-programming-interface-api).

| improve this answer | |
  • $\begingroup$ Thanks. There is indeed some information in the filenames. However, the CCLE cell line names are generally in the format {CELL_LINE_NAME}_{TISSUE_NAME}. For example, CJM_SKIN and CAL33_UPPER_AERODIGESTIVE_TRACT. I'm not sure where can I get the tissue names now. $\endgroup$ – rraadd88 Dec 11 '18 at 13:18
  • $\begingroup$ If you look at the "project" column in the above table, it gives the cancer type code from which the cell lines were extracted. Now how to map those abbreviations is less obious, but if you click a primary site filter in the left hand column, then you can work it out. $\endgroup$ – Ian Sudbery Dec 11 '18 at 13:31
  • 2
    $\begingroup$ This information is also present in the experimental metadata xml file attached to each bam file. This probably means its also accessible from the GDC metadata API. $\endgroup$ – Ian Sudbery Dec 11 '18 at 13:34
  • $\begingroup$ Ok. I see. There are XML files along with the BAM files. $\endgroup$ – rraadd88 Dec 11 '18 at 13:59
  • $\begingroup$ GDC API (gdc.cancer.gov/developers/…) should be the best way to dig into those. Thanks. $\endgroup$ – rraadd88 Dec 11 '18 at 13:59

I would recommend using UCSC Xena instead of GDC for CCLE data. They don't have everything. For example, BAM files are not available. However, the data that is available is organized in standard text files with consistent identifiers.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.