diff --git a/astrocyte_pkg.yml b/astrocyte_pkg.yml
index cf9d8a32de8c8824f89f4575c5d1a3818d1f0092..0f884de970f735e91f10cacfbb6777cb2b68495b 100644
--- a/astrocyte_pkg.yml
+++ b/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: |
diff --git a/workflow/configs/biohpc.config b/workflow/configs/biohpc.config
index 465d7e7bd09bbd7b6bcf8541c74fc87b271d0b73..9ff1e4bf38d24e68075bd6dbfd7719d75ca63741 100755
--- a/workflow/configs/biohpc.config
+++ b/workflow/configs/biohpc.config
@@ -1,6 +1,12 @@
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 = ''
+ // Below connects the experimental atac container to BioHPC's Slurm job scheduler.
+ // 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/'
+
// 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
@@ -16,7 +22,8 @@ process {
beforeScript = 'ulimit -Ss unlimited'
withName:runSource {
- container = 'docker://git.biohpc.swmed.edu:5050/s219741/astrocyte-atac-source/atac:0.0.1'
+ // Experimental containerized version of the caper software.
+ // container = 'docker://git.biohpc.swmed.edu:5050/s219741/astrocyte-atac-source/atac:0.0.1'
executor = 'local'
}
}
diff --git a/workflow/main.nf b/workflow/main.nf
index adc3f4bb718cd94cdd965e1f80caf985f165f3b2..7ebcf8035ecfd95091552ab9928c03fae85d27dc 100644
--- a/workflow/main.nf
+++ b/workflow/main.nf
@@ -1,8 +1,11 @@
/*
* 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 +26,43 @@ process runSource {
output:
file '*'
- """
- export LD_LIBRARY_PATH=/usr/lib64/:$LD_LIBRARY_PATH
+ shell:
+ '''
+ # Enable the use of bash-specific conda commands in this shell.
+ eval "$(conda shell.bash hook)"
+
+ module load python/3.8.x-anaconda
+ module load openjdk/18
+
+ # Create a temporary conda environment for caper.
+ conda create -y -c bioconda -c defaults -c conda-forge --name astrocyte-atac-caper python=3.8.18
+ # Activate and install caper.
+ conda activate astrocyte-atac-caper
+ export PATH=$PATH:~/.local/bin
+ pip install caper
+ # Creates a directory in the user's home2: ~/.caper/
+ caper init slurm
+
+ # 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 $inputJson --singularity --leader-job-name atac-source 1> batch_job.txt 2>> caper_err.txt
- """
+
+ # Launch the ATAC-seq leader job.
+ jobsubmit=$(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=$(bash !{baseDir}/scripts/checkJobState.sh "${jobsubmit}")
+ echo "Lead Job state check $(date) - State: $state" >> lead_job_check.txt
+ while [[ "$state" != *"COMPLETED"* ]] && [[ "$state" != *"FAILED"* ]] && [[ "$state" != *"CANCELLED"* ]]; do
+ sleep 15
+ state=$(bash !{baseDir}/scripts/checkJobState.sh "${jobsubmit}")
+ echo "Lead Job state check $(date) - State: $state" >> lead_job_check.txt
+ done
+
+ # Deactivate the temporary caper conda environment and delete it.
+ conda deactivate
+ conda remove --name astrocyte-atac-caper --all
+ rm -rf ~/.caper/
+ '''
}
diff --git a/workflow/scripts/checkJobState.sh b/workflow/scripts/checkJobState.sh
new file mode 100755
index 0000000000000000000000000000000000000000..d3e8d219222acf4679cb5aaa126ace232f91a804
--- /dev/null
+++ b/workflow/scripts/checkJobState.sh
@@ -0,0 +1,12 @@
+#!/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