From 06a445c3e7f12793d10b6cb29b3ba3d416099ea0 Mon Sep 17 00:00:00 2001
From: Chris Moth <chris.moth@vanderbilt.edu>
Date: Wed, 14 Jul 2021 11:53:18 -0500
Subject: [PATCH] Update clinvar SQL load slurm file to load per chromosome #25

---
 slurm/load_clinvar.slurm | 47 +++++++++++++++++++++++++++++++---------
 1 file changed, 37 insertions(+), 10 deletions(-)

diff --git a/slurm/load_clinvar.slurm b/slurm/load_clinvar.slurm
index a026584..5d06f3c 100755
--- a/slurm/load_clinvar.slurm
+++ b/slurm/load_clinvar.slurm
@@ -1,29 +1,56 @@
 #!/bin/sh
 #
-# Project        : PDBMap-v13 
+# Project        : PDBMap
 # Filename       : load_clinvar.slurm
-# Author         : R. Michael Sivley
+# Author         : R. Michael Sivley; updated Chris Moth June 2020
 # Organization   : Vanderbilt Genetics Institute,
 #                : Department of Biomedical Informatics,
 #                : Vanderbilt University
-# Email          : mike.sivley@vanderbilt.edu
-# Date           : 2015-12-28
+# Email          : chris.moth@vanderbilt.edu
+# Date           : 2021-07-09
+# Description    : Load each chromosome of clinvar variants separately, in parallel
 # Description    : Loads latest ClinVar release into PDBMap. 
 #                : sbatch load_clinvar.slurm
+# 
+# This script depends on the presence of (recently) downloaded clinvar .vcf files
 #=============================================================================#
-# Slurm Parameters
+# Slurm Parameters - tailor to your cluster, account, etc.
 #SBATCH -J Load_ClinVar
-#SBATCH --mail-user=mike.sivley@vanderbilt.edu
+#SBATCH --mail-user=chris.moth@vanderbilt.edu
 #SBATCH --mail-type=end
 #SBATCH --ntasks=1
 #SBATCH --time=5-0
-#SBATCH --mem=50GB
+#SBATCH --mem=32GB
 #SBATCH -o load_clinvar/load_clinvar.%A[%a].out
+#SBATCH -e load_clinvar/load_clinvar.%A[%a].err
+#SBATCH --array=1-24
 #=============================================================================#
 
-cd pdbmap # move into pdbmap main directory
-fname="/dors/capra_lab/data/clinvar/clinvar.vcf.gz"
+source  /dors/capra_lab/users/mothcw/psbadmin/psb_prep.bash
+cd $psbbin/../pdbmap # move into pdbmap main directory
+# fname="/dors/capra_lab/data/clinvar/clinvar.vcf.gz"
+fbase="/dors/capra_lab/data/clinvar/2021-07-07/GRCh38/clinvar_20210626.chr"
+ftail=".vcf"
 
+if (( ${SLURM_ARRAY_TASK_ID} > 1000000 )); then
+echo "You must start this slurm array script with --array=1-24"
+exit 1
+fi
+
+echo $warning
 # Syntax for autosomal chromosomes
-./pdbmap.py -c config/v13.config --dlabel=clinvar load_data ${fname}
+# Launch this script with sbatch --array=1-24 to populate the SLURM_ARRAY_TASK_ID
+chr=${SLURM_ARRAY_TASK_ID}
+
+if (( $chr == 23 )); then
+chr='X'
+elif (( $chr == 24 )); then
+chr='Y'
+fi
 
+
+# Syntax for autosomal chromosomes
+logfile="--logfile=slurm/load_clinvar/load_clinvar.${chr}.log"
+runcmd="./pdbmap.py -c $UDN/config/global.config ${logfile} load_vcf --dlabel clinvar ${fbase}${chr}${ftail}"
+echo "Executing $runcmd"
+`$runcmd`