Continue building up the workflow script using the container approach.

astrocyte_pkg.yml

@@ -13,12 +13,17 @@ author: 'Felix Perez, Achisha Saikia, Peng Lian'
 # A contact email address for questions
 email: 'felix.perez@utsouthwestern.edu, achisha.saikia@utsouthwestern.edu, biohpc-help@utsouthwestern.edu'
 # A more informative title for the workflow package
-title: 'ATAC-seq Source Workflow"
-
-This pipeline is designed for automated end-to-end quality control and processing of ATAC-seq. The pipeline can be run end-to-end, starting from raw FASTQ files all the way to peak calling and signal track generation using a single caper submit command. One can also start the pipeline from intermediate stages (for example, using alignment files as input). The pipeline supports both single-end and paired-end data as well as replicated or non-replicated datasets. The outputs produced by the pipeline include 1) formatted HTML reports that include quality control measures specifically designed for ATAC-seq and DNase-seq data, 2) analysis of reproducibility, 3) stringent and relaxed thresholding of peaks, 4) fold-enrichment and pvalue signal tracks.
+title: 'ATAC-seq Source Workflow'
 
 description: |
-  # TODO: Please describe the workflow. (AS)
+  This pipeline is designed for automated end-to-end quality control and processing of ATAC-seq.
+  The pipeline can be run end-to-end, starting from raw FASTQ files all the way to peak calling
+  and signal track generation using a single caper submit command. One can also start the pipeline from intermediate
+  stages (for example, using alignment files as input). The pipeline supports both single-end and paired-end data as
+  well as replicated or non-replicated datasets. The outputs produced by the pipeline include 1) formatted HTML reports
+  that include quality control measures specifically designed for ATAC-seq and DNase-seq data, 2) analysis of
+  reproducibility, 3) stringent and relaxed thresholding of peaks, 4) fold-enrichment and pvalue signal tracks.
+
 
 #### New Features in Astrocyte 0.4.0 and above ####
 citation: |

workflow/configs/biohpc.config

 singularity {
   enabled = true
-  runOptions = ' -B /cm/shared/apps/slurm/16.05.8 -B /etc/slurm -B /cm/shared/apps/slurm/var/etc/ -B /usr/lib64/libreadline.so.6 -B /usr/lib64/libhistory.so.6 -B /usr/lib64/libtinfo.so.5 -B /var/run/munge -B /usr/lib64/libmunge.so.2 -B /usr/lib64/libmunge.so.2.0.0 -B /cm/shared/apps/slurm/16.05.8/lib64/slurm/  -B /cm/shared/apps/slurm/16.05.8/lib64' 
+  runOptions = '\
+  --bind /cm/shared/apps/slurm/16.05.8,/etc/slurm,/cm/shared/apps/slurm/var/etc/,/usr/lib64/libreadline.so.6 \
+  --bind /usr/lib64/libhistory.so.6,/usr/lib64/libtinfo.so.5,/var/run/munge,/usr/lib64/libmunge.so.2 \
+  --bind /usr/lib64/libmunge.so.2.0.0,/cm/shared/apps/slurm/16.05.8/lib64/slurm/ \
+  --bind /cm/shared/apps/java/oracle/jdk1.8.0_231'
+
                                              // Please do NOT use "--disable-cache" in this runOptions. 
                                               // Starting from version 2.0.0, the astrocyte_cli will clean up the cache automatically.
   // runOptions = '--bind /vagrant:/vagrant'  // Use this one for vagrant development env only

workflow/main.nf

 /*
  * Copyright (c) 2024. The University of Texas Southwestern Medical Center
  *
- * TODO: (AC) Brief description of ATAC-seq (DONE)
- * ATAC-seq is a molecular biology technique that assesses chromatin accessibility in a genome. It uses a hyperactive Tn5 transposase to insert sequencing adapters into open chromatin regions, allowing researchers to identify and sequence these accessible genomic regions. ATAC-seq is widely used to study gene regulation, identify enhancers and promoters, and gain insights into chromatin structure.
+ * ATAC-seq is a molecular biology technique that assesses chromatin accessibility in a genome.
+ * It uses a hyperactive Tn5 transposase to insert sequencing adapters into open chromatin regions,
+ * allowing researchers to identify and sequence these accessible genomic regions. ATAC-seq is widely
+ * used to study gene regulation, identify enhancers and promoters, and gain insights into chromatin structure.
+ * 
  * @authors
  * Felix Perez, Achisha Saikia
  *
@@ -23,10 +26,35 @@ process runSource {
     output:
     file '*'
     
-    """
-    export LD_LIBRARY_PATH=/usr/lib64/:$LD_LIBRARY_PATH
+    shell:
+    '''
+    # Allow for the container to use the libraries & paths of Slurm on BioHPC.
+    export LD_LIBRARY_PATH=/atac/jdk-12/lib:/usr/lib64:/lib:$LD_LIBRARY_PATH
+    export PATH=/atac/jdk-12:/atac/jdk-12/bin:/bin:/cm/shared/apps/slurm/16.05.8/bin:$PATH
+
+    # Provide the container the SlurmUser (user and group) info used on Nucleus.
+    echo "slurm:x:450:450::/cm/local/apps/slurm:/bin/bash" >> /etc/passwd
+    echo "slurm:x:450:" >> /etc/group
+
+    # Source the container's entrypoint script to have access to the caper 
+    # commands to run the ATAC-seq pipeline in the runner.
+    source /atac/entrypoint.sh
+
+    # Record the relevant software versions.
+    java -version 2> java_version.txt
     sinfo -V > slurm_version.txt
     caper --version > caper_version.txt
-    caper hpc submit $baseDir/external_repo/astrocyte-atac-runner/atac.wdl -i $inputJson --singularity --leader-job-name atac-source 1> batch_job.txt 2>> caper_err.txt
-    """
+
+    # Launch the ATAC-seq leader job.
+    submit=$(caper hpc submit !{baseDir}/external_repo/astrocyte-atac-runner/atac.wdl -i !{inputJson} --singularity --leader-job-name atac-source)
+
+    # Monitor the state of the leader job; if it enters the COMPLETED, FAILED, or CANCELLED state, then finish the workflow process.
+    state=$(bash !{baseDir}/scripts/checkJobState.sh "${submit}")
+    echo "Lead Job state check $(date) - State: $state" >> lead_job_check.txt
+    while [[ "$state" != *"COMPLETED"* ]] && [[ "$state" != *"FAILED"* ]] && [[ "$state" != *"CANCELLED"* ]]; do
+        sleep 15
+        state=$(bash !{baseDir}/scripts/checkJobState.sh "${submit}")
+        echo "Lead Job state check $(date) - State: $state" >> lead_job_check.txt
+    done
+    '''
 }

workflow/scripts/checkJobState.sh

+#!/bin/bash
+
+# Get the jobID of the caper lead job from the input txt file.
+read -ra line <<< "$1"
+jobID=${line[3]}
+
+# Query Slurm for the state of the caper lead job.
+jobq=$(sacct --format State -j $jobID)
+
+# Return the caper lead job state.
+IFS=$'\n' read -rd '' -a jobstate <<< "$jobq"
+echo $(echo "${jobstate[2]}" | xargs)
\ No newline at end of file