As a test of runtimes, I ran one of my dockerized apps without an input file. The app has been written to exit immediately, in the event an input file is not specified, with an error message. On my local machine the app behaves as expected on being run without an input file - immediate exit with a help message. However on CGC, without an input file, the time it took to complete was 3 minutes. Of this , 2m and 14s was spent in the queue and 46 s was spent in running. My question is why did the app run for 46s before exiting? Exit should have been immediate. What is the overhead time being spent in? TIA for any insight. Anjan

Hi there, I'm trying to run a tool I developed. I committed the docker container and tested the execution of the tool locally. I get the following error when running my tool: "Error This task ran into a problem during execution and did not finish. Failed to start exmut. Internal error: Docker container failed to start." Is there anyway I can debug this? Thanks

Hello I have dockerized Picard in the following image: cgc-images.sbgenomics.com/purkayasthaa/runpicard-2.3.0. In brief, I use a perl script runPicard to run the "AddOrReplaceReadGroups" and "MarkDuplicates" sequentially, on a sorted bam file, within a single container. The image runs as expected on my local machine. However I get an error: "Failed to start runpicard. Internal error: Docker container failed to start. " on CGC. Any ideas on what may be going wrong? Thanks.

I'm developing a tool that uses multiple TCGA BAM and MAF files. I need to associate a mutation by its sample name to a BAM file or vice versa. Is there any metadata available to associate the BAM files to their sample name? Thanks in advance

I tried rerunning a task just now, and I almost immediately received the error: This task ran into a problem during execution and did not finish. Execution error occured <sic> I'm assuming this has to do with AWS? Or is it something with my account? It happened 3 times in a row. I'll keep trying every 30 minutes I guess.

What is the best way to redirect stderr in SBG? I see the option for stdout that adds a '>' to the command line, but what about when we want to add '2>'? Right now, I have the '2>' added as my last argument, but I get the error... This task ran into a problem during execution and did not finish. Command perl miRDeep2.pl /sbgenomics/Projects/5311483a-0df2-41d1-ba6e-361dc84bc31c/sample.reads.fa /sbgenomics/Projects/5311483a-0df2-41d1-ba6e-361dc84bc31c/hg19_bowtie.fa /sbgenomics/Projects/5311483a-0df2-41d1-ba6e-361dc84bc31c/sample.reads_vs_genome.arf /sbgenomics/Projects/5311483a-0df2-41d1-ba6e-361dc84bc31c/mature.fa none /sbgenomics/Projects/5311483a-0df2-41d1-ba6e-361dc84bc31c/hairpin.fa -P -t Human 2> report.log failed with exit code 2. ...and no error log, so I'm guessing that something is angry at my redirection. Is the shell bash? And if it is not, how do I log my stderr? Also, here is the documentation for the tool: https://www.mdc-berlin.de/36105849/en/research/research_teams/systems_biology_of_gene_regulatory_elements/projects/miRDeep/documentation#mirdeep2

I am running bowtie2 aligner imported from public apps to align RNAseq data (imported TCGA files) to indexed B37 (imported from common reference files) but again and again I am getting following error: Error This task ran into a problem during execution and did not finish. Command /opt/bowtie2-2.2.6/bowtie2 -p 8 -x GRCh37-lite_HPV_Redux --un 120815_UNC16-SN851_0177_BD13LJACXX_AGTTCC_L004_.unpaired_unaligned.fastq.gz --al 120815_UNC16-SN851_0177_BD13LJACXX_AGTTCC_L004_.unpaired_aligned.fastq.gz --un-conc 120815_UNC16-SN851_0177_BD13LJACXX_AGTTCC_L004_.paired_unaligned.fastq.gz -a --rg-id id --rg PL:Illumina_HiSeq --rg SM:TCGA-CQ-6229-01A -1 /sbgenomics/Projects/5eeeb0e9-950d-42dc-9570-2eb54cee03a5/120815_UNC16-SN851_0177_BD13LJACXX_AGTTCC_L004_1.fastq -2 /sbgenomics/Projects/5eeeb0e9-950d-42dc-9570-2eb54cee03a5/120815_UNC16-SN851_0177_BD13LJACXX_AGTTCC_L004_2.fastq -S ./120815_UNC16-SN851_0177_BD13LJACXX_AGTTCC_L004.sam && mv 120815_UNC16-SN851_0177_BD13LJACXX_AGTTCC_L004_.paired_unaligned.fastq.1.gz 120815_UNC16-SN851_0177_BD13LJACXX_AGTTCC_L004_.paired_unaligned.1.fastq.gz && mv 120815_UNC16-SN851_0177_BD13LJACXX_AGTTCC_L004_.paired_unaligned.fastq.2.gz 120815_UNC16-SN851_0177_BD13LJACXX_AGTTCC_L004_.paired_unaligned.2.fastq.gz failed with exit code 137. Can anybody suggest what is going wrong? I searched about error 137 but didn't find proper informaiton. Someone has suggested that 137 indicates that process is killed by some other process or by someone, but I don't understand why would that happen? Thanking you - Pratik

I am wrapping a tool and enter my docker repository (miasteinberg/mirdeep2:0.0.7) in the repository entry field. I click on the Save button. I get the green box saying my tool was successfully updated, but the Docker Repo field is empty now. I even copied and pasted the repo from a previous tool that definitely works to make sure it wasn't a typo. Same thing. After saving, the field becomes blank. I restarted Firefox - same problem. I tried Chrome - same problem. What should I try next?

Hi, Please explain the use of $self in the Value field of a job input spec. I don't know when and how I can use it. For example, say the ID of my input parameter is called 'quality'. In the JavaScript expression I want to enter in the Value field, it seems I must refer to the parameter as $job.inputs.quality, but I thought I could also do $self.quality. In the documentation at http://docs.cancergenomicscloud.org/docs/dynamic-expressions-in-tool-descriptions#section-the-self-object, it says: """ In the command line bindings for the tool inputs, $self is set to the value of the input. In other words, in this context $self is just $job.inputs.<input_id>. """ According to that documentation, I should be able to use an expression such as "$self || 5" to create a default value of 5. But when doing so, I get this error when trying to run my tool: """ Error This task ran into a problem during execution and did not finish. Failed to start trimgalore. Failed to build command line. Failed to build command line. Failed evaluating expression $self || 20. org.mozilla.javascript.EcmaError: ReferenceError: "$self" is not defined. (script#1). """ So why does it say that $self is not defined?

I am repeatedly getting errors while trying to upload this 31GB file: [[email protected] login_node ~/software/cgc-uploader]$ ls -lh /scratch/users/nathankw/IndelIndices.tar.gz -rw-r--r-- 1 nathankw ruthm 31G May 13 22:13 /scratch/users/nathankw/IndelIndices.tar.gz [[email protected] login_node ~/software/cgc-uploader]$ ./bin/cgc-uploader.sh -p 0b57a269-388b-4a99-b197-fd9206a00375 /scratch/users/nathankw/IndelIndices.tar.gz Initializing upload... Starting upload of 1 file(s) to 'salzman' 5736ddbfe4b0e5be418274b4 /scratch/users/nathankw/IndelIndices.tar.gz 100.00% Upload failed (FAILED) When I look at the project on CGC that I was uploading to, it looks like the file is all there size-wise. Why do I get the failed error message?

Dear CGC team, When I compare the mutation data I download from the TCGA website (maf files), they don't match up with the data displayed in CGC. For example, in ovarian cancer 8% and 9% of patients are BRCA1 and BRCA2 mutated. When I pull these data up in the browser, there are only 2 BRCA2 mutations showing up. The same pattern is true for TCGA LUSC and BLCA, with CGC displaying fewer mutations than shown in the raw data. Can you please explain this discrepancy? Thanks, An

When I try to run `samtools mpileup` it expects to find the BAM index in the same folder as the BAM file with the extension ".bai". I'm not sure how I can expose the BAM index to a tool that assumes a certain file location for the index.

I'd like to test some of my tools on this platform. I was wondering if there was any data that's not TCGA-restricted that everyone can access and use, just for getting familiar with the environment. This could perhaps be 1kG data or cell-line data. I guess the alternative is to upload and use private data but I wanted to check what's recommended. Thanks, Avi

My program reads and writes files from a directory but the CGC doesn’t have a folder structure. How can I use the SDK to help run my app on the CGC?

I'm new to Docker. How can I get started? Any helpful tips?

I use my eRA Commons ID to log into the CGC. How do I use the Docker CLI to log in and push images to the CGC image registry? What do I need?

Is there a database available to query for the TCGA bam files?

I've joined today as a beta user and started a project, but once I do a query and pull up a list of files, I'm unable to highlight them either on the query map or the list of files in order to add them to my project for analysis. Is this a bug or am I doing something wrong? The help page indicates that I should be able to click on a path to highlight and add those files. Thanks!