how actually does one run your atac pipeline? super frustrated

Hi, Glad to see that there is this ENCODE ATAC-seq pipeline. We are using some HPC to perform the computing, and as most others, we are not the admin, so a lot of times very difficult to install those packages. Anyway, I ve git clone your ‘atac-seq-pipeline’ files to folder with the same name on our HPC. We have also git cloned the Caper. I modified your minimum json input specification file to this [input.json]:

{
    "atac.pipeline_type" : "atac",
    "atac.genome_tsv" : "https://storage.googleapis.com/encode-pipeline-genome-data/mm10_caper.tsv",
    "atac.fastqs_rep1_R1" : [
        "primary_seq/repa_1.fastq.gz",
        "primary_seq/repb_1.fastq.gz",
    ],
    "atac.fastqs_rep1_R2" : [
        "primary_seq/repa_2.fastq.gz",
        "primary_seq/repb_2.fastq.gz",
    ],
    "atac.paired_end" : true,
    "atac.multimapping" : 4,
    "atac.auto_detect_adapter" : true,
    "atac.trim_adapter.cpu" : 2,
    "atac.bowtie2_cpu" : 8,
    "atac.bowtie2_mem_mb" : 16000,
    "atac.filter_cpu" : 2,
    "atac.filter_mem_mb" : 12000,
    "atac.macs2_mem_mb" : 16000,
    "atac.smooth_win" : 73,
    "atac.enable_idr" : true,
    "atac.idr_thresh" : 0.05,
    "atac.enable_xcor" : true,
    "atac.title" : "Project1",
    "atac.description" : "Project1"
}

And then I submitted a job running this line: caper run atac.wdl -i $outdir/input.json --deepcopy --use-singularity and I got this error immediately:

usage: caper run [-h] [--dry-run] [-i INPUTS] [-o OPTIONS] [-l LABELS]
                 [-p IMPORTS] [-s STR_LABEL] [--hold]
                 [--singularity-cachedir SINGULARITY_CACHEDIR] [--no-deepcopy]
                 [--deepcopy-ext DEEPCOPY_EXT]
                 [--docker [DOCKER [DOCKER ...]]]
                 [--singularity [SINGULARITY [SINGULARITY ...]]]
                 [--no-build-singularity] [--slurm-partition SLURM_PARTITION]
                 [--slurm-account SLURM_ACCOUNT]
                 [--slurm-extra-param SLURM_EXTRA_PARAM] [--sge-pe SGE_PE]
                 [--sge-queue SGE_QUEUE] [--sge-extra-param SGE_EXTRA_PARAM]
                 [--pbs-queue PBS_QUEUE] [--pbs-extra-param PBS_EXTRA_PARAM]
                 [-m METADATA_OUTPUT] [--java-heap-run JAVA_HEAP_RUN]
                 [--db-timeout DB_TIMEOUT] [--file-db FILE_DB] [--no-file-db]
                 [--mysql-db-ip MYSQL_DB_IP] [--mysql-db-port MYSQL_DB_PORT]
                 [--mysql-db-user MYSQL_DB_USER]
                 [--mysql-db-password MYSQL_DB_PASSWORD] [--cromwell CROMWELL]
                 [--max-concurrent-tasks MAX_CONCURRENT_TASKS]
                 [--max-concurrent-workflows MAX_CONCURRENT_WORKFLOWS]
                 [--max-retries MAX_RETRIES] [--disable-call-caching]
                 [--backend-file BACKEND_FILE] [--out-dir OUT_DIR]
                 [--tmp-dir TMP_DIR] [--gcp-prj GCP_PRJ]
                 [--gcp-zones GCP_ZONES] [--out-gcs-bucket OUT_GCS_BUCKET]
                 [--tmp-gcs-bucket TMP_GCS_BUCKET]
                 [--aws-batch-arn AWS_BATCH_ARN] [--aws-region AWS_REGION]
                 [--out-s3-bucket OUT_S3_BUCKET]
                 [--tmp-s3-bucket TMP_S3_BUCKET] [--use-gsutil-over-aws-s3]
                 [-b BACKEND] [--http-user HTTP_USER]
                 [--http-password HTTP_PASSWORD] [--use-netrc]
                 wdl
caper run: error: argument --deepcopy-ext: expected one argument

What seems to be the problem? Seems the caper run options are different from your documentations on the landing page of this project?

We have “python3/3.6.4” and “singularity/2.5.2” loaded in out HPC.

[akundaje]

That’s not how the installation works. All dependencies are automatically installed. Jin will reply with the correct way to install the pipeline.

[leepc12]

We will make a release with fixed documentation soon this week. Please remove --deepcopy from the command line and use --singularity instead of --use-singularity.