How to find TCGA data in cghub?

[lethalfang]

We have gotten access approval for some TCGA data, but how do I find them? I have GeneTorrent (gtdownload and cgquery) installed, but it seems mightily difficult to find anything I'm looking for.

For instance, I'm trying to download a couple of data sets from TCGA's lung adenocarcinoma studies: http://www.sciencedirect.com/science...92867412010616

"The dbGAP accession number for the data reported in this paper is phs000488.v1.p1."

The dbGAP page can be found here: http://www.ncbi.nlm.nih.gov/projects...hs000488.v1.p1

cgquery "study=phs000488" returned zero result, as is the case for pretty much all the accession numbers I've found in any paper.

I downloaded some supplementary files from the article's website, but couldn't identify any of the Patient ID in the cghub's data manifest file, e.g., http://www.broadinstitute.org/pubs/l...UAD-5V8LT.html


So....... does anyone know actually how to find the data set you're looking for?

Thanks.

[Richard Finney]

FOR BAM FILES ...

Have you seen this ...

https://browser.cghub.ucsc.edu/

I usually don't like these JavaScript GUI click click click things, but this is not so bad.

You want the analysis_ids.

My script for this (which you must customize to your location ) is ..

#point to your executable and libraries for cghub client where you put them
export LD_LIBRARY_PATH=/data/data04/CG/cghub/lib/:/data/data04/CG/cghub/lib/GeneTorrent/:/h1/finneyr/xerces-c-3.0.1/src/.libs/:/h1/finneyr/XQilla-2.2.3/.libs/:LD_LIBRARY_PATH
export PATH=/data/data04/CG/cghub/bin/:$PATH:

function f
{
gtdownload -vv -c /h1/finneyr/finneyr.key -d $1
sleep 2;
}

# just add "f analysis_id"

f 038d680d-4a29-4be1-9568-72d80a52c782
f 059e80af-c614-4424-8075-d42f072705b2



ALTERNATELY
You can grab info for BAMs for a project like this ...
function f
{
echo $1
cghub/bin/cgquery disease_abbr=$1 > cghub.$1.txt
n=$((i%5))
if [ $n -eq 0 ]; then sleep 1; fi
((i=i+1))
sleep 2;
}

f LIHC
f LUAD

This creates reports for LIHC (liver) and LUAD (lung adeno).
You can parse out the analysis_ids.



FOR OTHER STUFF ...

http://tcga-data.nci.nih.gov///tcgafiles/ftp_auth/distro_ftpusers/anonymous/

FOR PROTECTED OTHER STUFF ...

https://tcga-data-secure.nci.nih.gov///tcgafiles/tcga4yeo/

(you need to log in)

For the tcga-data.nci.nih.gov sites , you can write a script to grab a listing of all files.

[GenePool]

Hi,

Yep, like Richard has pointed out, you need to get a hold of analysis IDs.

We typically use https://browser.cghub.ucsc.edu/ to search for the samples that we're interested in, and then link the samples to the patient & sample metadata available for the various cohorts here: https://tcga-data.nci.nih.gov/tcgafi...onymous/tumor/

For what it's worth, Station X has spent a lot of timing organizing and curating the patient & sample metadata, and subsequently attaching to the various genomics assays generated by The Cancer Genome Atlas. This data is all prepped and ready for analysis in GenePool.

If you're interested, here are some related posts about it:

http://seqanswers.com/forums/showthread.php?t=48485

http://seqanswers.com/forums/showthread.php?t=42471

Good Luck!

------------------------------
GenePool is making genomics data management, analysis, and sharing easier!
Products @ www.stationxinc.com