Gistic2 reproducibility

[Petri Pölönen]

Dear cbioportal team,

I downloaded the dataset for TARGET ALL (all_phase2_target_2018_pub.tar.gz). I would like to reproduce the gistic results using file: all_phase2_target_2018_pub_data_cna_hg19.seg.

Unfortunately, I encounter problem, as the downloaded set of files do not contain marker (-mk) file. I tried to use file from broad institute (genome.info.6.0_hg19.na31_minus_frequent_nan_probes_sorted_2.1.txt) but I still encounter error that markers do not match the all_phase2_target_2018_pub_data_cna_hg19.seg file.

It would be great if you could send me the command how you run Gistic2 for this dataset, and/or provide me the correct marker file to run gistic2 myself.

Thanks for all the effort with cbioportal, it is truly great work!

Petri Pölönen, PhD student,
University of Eastern Finland
Biomedicine

[Ethan Cerami]

Petri,

We don’t actually run GISTIC ourselves as part of the cBioPortal
pre-processing pipeline.  Rather, for TCGA data, we ingest the GISTIC
results from the Broad Firehose platform.

You can find full information here:  https://gdac.broadinstitute.org/

Ethan

--
Ethan Cerami

On October 31, 2018 at 9:47:04 AM, Petri Pölönen (petri....@uef.fi) wrote:
> > Petri

[Petri Pölönen]

Thank you for a promt reply,

TARGET data is not from the TCGA, so I was wondering if the processing was done via TARGET team or also processed at broad. I used the same marker file as firehose broad processing does, but it did not work.

Best,
Petri

[Ethan Cerami]

Not sure about that, but we are checking and will get back to you...

--
Ethan Cerami

On October 31, 2018 at 11:18:18 AM, Petri Pölönen (petri....@uef.fi) wrote:
> Thank you for a promt reply,
>
> TARGET data is not from the TCGA, so I was wondering if the processing was done via TARGET
> team or also processed at broad. I used the same marker file as firehose broad processing
> does, but it did not work.
>
> Best,
> Petri
>

[aocho...@gmail.com]

Hi Petri, 

Thank you for your patience. Please follow the link below to download the marker file used for the TARGET-ALL P2 GISTIC run. Let us know if you have any additional questions. 

Link: https://drive.google.com/file/d/1mcUhc6DBRsLsN_uNMx_DNnZsmQnbetrN/view?usp=sharing 

Best,

Angelica 

[Nikolaus Schultz]

Thanks Angelica.

Adding Petri to the cc to make sure the message gets through.

-- 
You received this message because you are subscribed to the Google Groups "cBioPortal for Cancer Genomics Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cbioportal+...@googlegroups.com.
To post to this group, send email to cbiop...@googlegroups.com.
Visit this group at https://groups.google.com/group/cbioportal.
To view this discussion on the web visit https://groups.google.com/d/msgid/cbioportal/46f5eca3-1442-441c-8375-579b4674bfdb%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Petri Pölönen]

Thank you,

I have downloaded the file. I appreciate the help!

Petri

On 31 Oct 2018, at 12.25, Nikolaus Schultz <nsch...@gmail.com> wrote:

https://drive.google.com/file/d/1mcUhc6DBRsLsN_uNMx_DNnZsmQnbetrN/view?usp=sharing 

[Petri Pölönen]

Hi,

I was not able to replicate the results from “data_CNA.txt” available in “all_phase2_target_2018_pub.tar.gz”, I’m assuming I should get same result to gistic output “all_thresholded.by_genes.txt". Could you point out how you run gistic2 (inputs and parameters)?

My run:

Gistic annotation file:

hg19.UCSC.add_miR.140312.refgene.mat

Seq file:

all_phase2_target_2018_pub_data_cna_hg19.seg # file from all_phase2_target_2018_pub.tar.gz. already log2()-1 as required by gistic2

Markers file (from cbioportal team)

markers-all.txt

Parameters:

gistic2 -b $basedir -seg $segfile -mk $markersfile -refgene $refgenefile -ta 0.1 -td 0.1 -qvt 0.25 -cap 1.5 -rx 0 -js 4 -maxseq 2000 -genegistic 1 -smallmem 1 -broad 1 -brlen 0.7 -conf 0.99 -armpeel 1 -savegene 1 -gcm extreme

Best,

Petri

On 31 Oct 2018, at 12.25, Nikolaus Schultz <nsch...@gmail.com> wrote:

[Angelica]

Hi Petri, 

Here is the command used

gistic -b $basedir -seg $segfile -mk $markersfile -refgene $refgenefile -genegistic 1 -ta 0.1 -td 0.1 -qvt 0.25 -maxseg 2000 -cap 1.5 -brlen 0.5 -rx 0 -conf 0.95 -js 8 -armpeel 1 -savegene 1

Angelica

On Wednesday, October 31, 2018 at 9:47:04 AM UTC-4, Petri Pölönen wrote: