Help with WDL workflow for mutec2?
Answered
Hi this is concerning the workflow: https://github.com/gatk-workflows/gatk4-somatic-snvs-indels
Not sure where to ask but I thought I try here first. I'm trying to run the example in the json locally. All files have been donwloaded locally and using cromwell-49.jar. However the last merging step seem to have this error I cannot dicipher and wondering if someone can help or point me to the right direction?
the error is this:
cromwell.backend.standard.StandardAsyncExecutionActor$$anon$2: Failed to evaluate job outputs:
Bad output 'M2.tumor_sample': Failed to read_string("tumor_name.txt") (reason 1 of 1): Futures timed out after [60 seconds]
Bad output 'M2.normal_sample': Failed to read_string("normal_name.txt") (reason 1 of 1): Futures timed out after [60 seconds]
at cromwell.backend.standard.StandardAsyncExecutionActor.$anonfun$handleExecutionSuccess$1(StandardAsyncExecutionActor.scala:916)
at scala.util.Success.$anonfun$map$1(Try.scala:255)
at scala.util.Success.map(Try.scala:213)
so its strange because the "tumor_name.txt" is literally a string and in each sub directory I can clearly see that there are these files. Anyone know what is going on here? thanks.
A
-
In the WDL script the output block for the M2 task specifies two variables
String tumor_sample = read_string("tumor_name.txt")
String normal_sample = read_string("normal_name.txt")The read_string() function should be reading the contents of those files, perhaps the files are empty? Try checking the log.stderr and log.stdout files to confirm M2 task ran without any problems.
-
Also, the repo you're pointing to is archived, it's best to use the latest version of the workflow found here.
-
Beri thanks. What is strange is that each subdirectory looks fine. There are no errors. and both tumor_name.txt and normal has a text to the file names. Each subdirectory has a a output vcf. Stdout looks fine
Tool returned:
SUCCESS
Tool returned:
SUCCESS
Tool returned:
SUCCESSMoreover the stderr no errors, the last entry was a command.
Running:
java -Dsamjdk.use_async_io_read_samtools=false -Dsamjdk.use_async_io_write_samtools=true -Dsamjdk.use_async_io_write_tribble=false -Dsamjdk.compression_level=2 -Xmx3000m -jar /root/gatk.jar GetPileupSummaries -R /cromwell-executions/Mutect2/40fdbfbc-bc9c-4bdf-be88-3630342aaca1/call-M2/shard-40/inputs/-1275058761/Homo_sapiens_assembly19.fasta -I /cromwell-executions/Mutect2/40fdbfbc-bc9c-4bdf-be88-3630342aaca1/call-M2/shard-40/inputs/2119282986/HM5086F_A.b37.bam --interval-set-rule INTERSECTION -L /cromwell-executions/Mutect2/40fdbfbc-bc9c-4bdf-be88-3630342aaca1/call-M2/shard-40/inputs/1020858285/0040-scattered.interval_list -V /cromwell-executions/Mutect2/40fdbfbc-bc9c-4bdf-be88-3630342aaca1/call-M2/shard-40/inputs/-1275058761/small_exac_common_3.vcf -L /cromwell-executions/Mutect2/40fdbfbc-bc9c-4bdf-be88-3630342aaca1/call-M2/shard-40/inputs/-1275058761/small_exac_common_3.vcf -O normal-pileups.table -
So If there aren't any errors in the stderr and GATK says it was successful chances are it related to Cromwell/WDL.
The WDL documentation has some info mentioning possible issues when using read_string()If the entire contents of the file can not be read for any reason, the calling task or workflow will be considered to have failed. Examples of failure include but are not limted to not having access to the file, resource limitations (e.g. memory) when reading the file, and implementation imposed file size limits.
-
Beri thanks. What is bizaare is that even if I hard coded the file names, what just get rid of read_string() then there are other errors. I've just pretty much given up on this since it would've been faster had I just build this from scratch!
-
Hi,
Was there any solution to this error? I am getting the same error when I tried to run this Mutect2 wdl using cromwell on my server - I imagine it has to do with runtime attributes you have to change when it isn't used on terra as this workflow is working when I use it on terra vs. on my institute's cluster (same samples/runtime attributes are giving me this error). Thank you!
Toby
-
Hi TA
Can you post the actual error messages you get and also provide versions of tools that you use to run these workflows?
We can help you better with that information since this topic is already 3 years old and there has been many updates to all tools mentioned here.
Regards.
-
Hi,
The error message is:
cromwell.backend.standard.StandardAsyncExecutionActor$$anon$2: Failed to evaluate job outputs:
Bad output 'M2.tumor_sample': Failed to read_string("tumor_name.txt") (reason 1 of 1): Future timed out after [60 seconds]
Bad output 'M2.normal_sample': Failed to read_string("normal_name.txt") (reason 1 of 1): Future timed out after [60 seconds]
at cromwell.backend.standard.StandardAsyncExecutionActor.$anonfun$handleExecutionSuccess$1(StandardAsyncExecutionActor.scala:990)
at scala.concurrent.impl.Promise$Transformation.run(Promise.scala:467)
at akka.dispatch.TaskInvocation.run(AbstractDispatcher.scala:41)
at akka.dispatch.ForkJoinExecutorConfigurator$AkkaForkJoinTask.exec(ForkJoinExecutorConfigurator.scala:49)
at akka.dispatch.forkjoin.ForkJoinTask.doExec(ForkJoinTask.java:260)
at akka.dispatch.forkjoin.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1339)
at akka.dispatch.forkjoin.ForkJoinPool.runWorker(ForkJoinPool.java:1979)
at akka.dispatch.forkjoin.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:107)
The tools I am using are all the tools from the most recent gatk docker container:
us.gcr.io/broad-gatk/gatk@sha256:53b4590c3c8d2d5db7a02c07754a69ea134841f6a08a5a16b238bbf2d6ebd250
Basically I am running best practices mutect2 wdl here:
https://github.com/broadinstitute/gatk/blob/master/scripts/mutect2_wdl/mutect2.wdl
And it is working up until the end of the mutect2 task. I think there is some issue with scatter-gather when it is implemented locally/on a cluster with slurm as opposed to terra with runtime parameters but not exactly sure what runtime parameters to change. We implemented cromwell on the server following the cromwell instructions here:
https://cromwell.readthedocs.io/en/stable/backends/HPC/
-
Hi TA
Can you give us more details about what version of cromwell you are using? Also can you provide us the cromwell config file for this installation so that we can debug the issue?
Regards.
Please sign in to leave a comment.
9 comments