Can I install this ATAC-pipeline in Mac? #132
Comments
|
You can try removing trim-galore and change the version of bowtie2 in However, we strongly recommend to install docker and use our new pipeline which supports docker https://github.com/ENCODE-DCC/atac-seq-pipeline/. |
|
Hi Jin, Thank you for your suggestion! I was trying to install the new ENCODE-ATAC-seq-pipeline on my local computer with docker yesterday. But when I ran the pipeline, it seems the pipeline was stuck in "bowtie2" step. The computer keep running till now and seems will run forever and become very hot. Here is the input.json: } Here is the hg38.tsv ref_fa /Users/liangzhu/data/encode-pipeline-genome-data/hg38/GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta.gz Here is the details of running progress: Liangzhus-MacBook-Pro:atac-seq-pipeline-master liangzhu$ java -jar -Dconfig.file=backends/backend.conf cromwell-33.1.jar run atac.wdl -i input.json -o workflow_opts/docker.json run as in the original configuration without --rm flag (will remove later)docker run get the return code (working even if the container was detached)rc=$(docker wait cat /Users/liangzhu/atac-seq-pipeline-master/cromwell-executions/atac/e0ae37ea-6dc1-471e-825a-99d79d886bad/call-read_genome_tsv/execution/docker_cid) remove the container after waitingdocker rm cat /Users/liangzhu/atac-seq-pipeline-master/cromwell-executions/atac/e0ae37ea-6dc1-471e-825a-99d79d886bad/call-read_genome_tsv/execution/docker_cid return exit codeexit $rc run as in the original configuration without --rm flag (will remove later)docker run get the return code (working even if the container was detached)rc=$(docker wait cat /Users/liangzhu/atac-seq-pipeline-master/cromwell-executions/atac/e0ae37ea-6dc1-471e-825a-99d79d886bad/call-trim_adapter/shard-0/execution/docker_cid) remove the container after waitingdocker rm cat /Users/liangzhu/atac-seq-pipeline-master/cromwell-executions/atac/e0ae37ea-6dc1-471e-825a-99d79d886bad/call-trim_adapter/shard-0/execution/docker_cid return exit codeexit $rc run as in the original configuration without --rm flag (will remove later)docker run get the return code (working even if the container was detached)rc=$(docker wait cat /Users/liangzhu/atac-seq-pipeline-master/cromwell-executions/atac/e0ae37ea-6dc1-471e-825a-99d79d886bad/call-bowtie2/shard-0/execution/docker_cid) remove the container after waitingdocker rm cat /Users/liangzhu/atac-seq-pipeline-master/cromwell-executions/atac/e0ae37ea-6dc1-471e-825a-99d79d886bad/call-bowtie2/shard-0/execution/docker_cid return exit codeexit $rc The pipeline is stuck on the last step and keep running. Is there any problem? |
|
MacBook Pro doesn't have enough resources to run pipelines smoothly. Please try to add |
|
Thank you, I will try. |
|
Hi Jin, I am trying to run the pipeline on google platform and DNANexus. I met problems on both platform.
In fact, I created a bucket named 'atacout' in my google project, shall I create a file under such directory? How?
the error message is as below: Do I need to modify the docker.json file? or is there any problems in my input.json? } Thanks a lot! |
|
It's a Google Cloud bucket permission problem. Go to the Google Cloud Project console and click on the vertical Also, don't forget to log in before running pipelines. |
|
Hi Jin, java -jar -Dconfig.file=backends/backend.conf -Dbackend.default=google -Dbackend.provider.google.config.project=liangzhu-atac2018 -Dbackend.provider.google.config.root=atacout cromwell-33.1.jar run atac.wdl -i input-google.json -o workflow_opts/docker.json Below is the total information of process: Caused by: com.google.api.client.http.HttpResponseException: 403 Forbidden [2018-07-16 16:43:49,69] [error] WorkflowManagerActor Workflow e6b20ecd-5059-4c69-b624-28bab6740e18 failed (during ExecutingWorkflowState): cromwell.core.CromwellFatalException: com.google.api.client.http.HttpResponseException: 403 Forbidden Caused by: com.google.api.client.http.HttpResponseException: 403 Forbidden [2018-07-16 16:43:49,69] [info] WorkflowManagerActor WorkflowActor-e6b20ecd-5059-4c69-b624-28bab6740e18 is in a terminal state: WorkflowFailedState |
|
Hi Jin,
when I run the pipeline on google platform, below information appeared: [2018-07-17 10:17:08,43] [info] PipelinesApiAsyncBackendJobExecutionActor [8cd46ebfatac.read_genome_tsv:NA:1]: echo "Reading genome_tsv /cromwell_root/encode-pipeline-genome-data-1/hg38_google.tsv ..." So what I am confusing is the path:"Reading genome_tsv /cromwell_root/encode-pipeline-genome-data-1/hg38_google.tsv ...". The "encode-pipeline-genome-data-1/hg38_google.tsv" is the correct path for my project, but the "Reading genome_tsv /cromwell_root/" is not. There is no such buckets in my project. Is it the reason that forbid my access to the bucket? I am really exhausted about it...... Thank you for your patience. |
|
Please try with You don't need to make your own genome data buckets. Pick a genome TSV file under |
|
I tried and picked a genome TSV file under gs://encode-pipeline-genome-data/ in the input.json. But it still doesn't work..... I used the command line gsutil iam ch -d user:lxl623@case.edu:objectCreator,objectViewer gs://my-bucket . The command successfully added access permission to the bucket. I don't know why I couldn't access the bucket (even your public bucket gs://encode-pipeline-genome-data/ ) when I was running the pipeline on google platform. Shall I install a cromwell server on Google platform as showed in the test_workflow? Fortunately, I am able to run the pipeline on the DNAnexus now. At the very beginning, I used the latest version dxWDL0.69, but it didn't work until I replaced it with the old version dxWDL-0.66.1.jar. |
|
@llz-hiv, that dxWDL thing is a known issue. For the latest dxWDL, you need to add About the bucket permission problem... Does |
|
Hi Jin, I tried the command gsutily cp.... |
|
As described in the instruction web page, did you run the following to store Google Cloud credential information on your computer? |
|
FYI, I created a new google account and tested and it worked fine... |
|
Hi Jin, I'm trying to run the pipeline locally on a Mac, but libgcc 5.2 is not available for OSX. I installed the pipeline with Docker to get around this issue, but now I can't build genome database because that requires me to install_dependencies.sh, which gives me the error about libgcc=5.2.0 is not available. Is there a way to work around this? Could I use a genome downloaded from somewhere? (I'm working with mm10.) Thanks! |
I installed Miniconda3-latest-MacOSX-x86_64.sh, Bigdata script (bds_Darwin.tar).
Also delete anaconda in my Mac and removed Anaconda path from .bash_profile.
But when I run "bash install_dependencies.sh", it still use anaconda. the detail as below:
Solving environment: failed
PackagesNotFoundError: The following packages are not available from current channels:
Current channels:
To search for alternate channels that may provide the conda package you're
looking for, navigate to
and use the search bar at the top of the page.
how to fix it? It seems that I can't delete anaconda completely.
Ps: I use the method showed in this link (https://docs.anaconda.com/anaconda/install/uninstall) to delete anaconda.
The text was updated successfully, but these errors were encountered: