updated HPC scripts

scripts/create_jobs.sh

@@ -1,20 +1,25 @@
 #!/usr/bin/bash                                                                 
+if [ -z "$1" ]; then 
+	echo "Provide cutoff"
+	exit 1
+fi
+cutoff=$1
 
-if [[ $1 = "--rerun" ]];                                                        
+if [[ $2 = "--rerun" ]];                                                        
 then                                                                            
     echo "submitting rerun jobs"                                                
     #./get_rerun_residues.py                                                    
     IFS=,                                                                       
-    read -r -a residues < rerun_residues.csv                                    
+    read -r -a residues < rerun_residues_$cutoff.csv                                    
 else                                                                            
     echo "submitting all jobs"                                                  
     IFS=,                                                                       
-    read -r -a residues < residue_list.csv                                      
+    read -r -a residues < residue_list_$cutoff.csv                                      
 fi                                                                              
 
 
 for res in ${residues[@]}; do                                                   
-    sed -r "s/RESIDUE/${res}/g;s/RESID/${res:1}/g" submit_tmp.slu > submit.slu;
+    sed -r "s/RESIDUE/${res}/g;s/RESID/${res:1}/g;s/CUTOFF/$cutoff/g" submit_tmp.slu > submit.slu;
     sbatch submit.slu                                                           
     done                                                                        
 

scripts/get_rerun_residues.py

@@ -1,7 +1,12 @@
 from glob import glob                                    
 import numpy as np
 import os                                                
+import sys
 
+if len(sys.argv) < 2:
+    sys.exit('Provide cutoff')
+
+cutoff = sys.argv[1]
 bashCommand = "sacct -n -X --format jobname,jobid -s RUNNING > .running_jobs.txt"
 os.system(bashCommand)
 
@@ -10,7 +15,7 @@
     for line in f:
         running.append(line.strip().split()[0])
 
-os.chdir('basicrta-7.0/')
+os.chdir(f'basicrta-{cutoff}/')
 dirs = np.array(glob('?[0-9]*'))
 
 if len(running) > 0:
@@ -28,7 +33,7 @@
         rerundirs.append(adir)
 
 rerunids = [adir for adir in rerundirs]
-with open('../rerun_residues.csv', 'w') as w:
+with open(f'../rerun_residues_{cutoff}.csv', 'w') as w:
     for res in rerunids:
         w.write(f'{res},') 
 

scripts/setup.py

@@ -3,28 +3,39 @@
 import MDAnalysis as mda
 import pickle
 
-with open('contacts_7.0.pkl', 'rb') as f:
-    a = pickle.load(f)
+if __name__=="__main__":
+    import argparse
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--contacts', type=str)
+    args = parser.parse_args()
 
-rg = a.dtype.metadata['ag1'].residues 
-ids = rg.resids
-names = rg.resnames
-names = np.array([mda.lib.util.convert_aa_code(name) for name in names])
-uniqs = np.unique(a[:, 0]).astype(int)
+    dirname = args.contacts.strip('.pkl')
+    cutoff = dirname.split('_')[1]
 
-inds = np.array([np.where(ids==val)[0][0] for val in uniqs])
-resids, resnames = ids[inds], names[inds]
-residues = np.array([f'{name}{resid}' for name, resid in zip(resnames, resids)])
+    with open(f'{args.contacts}', 'rb') as f:
+        a = pickle.load(f)
 
-with open('residue_list.csv', 'w+') as r:
-    for residue in residues:
-        r.write(f'{residue},')
+    rg = a.dtype.metadata['ag1'].residues 
+    ids = rg.resids
+    names = rg.resnames
+    names = np.array([mda.lib.util.convert_aa_code(name) for name in names])
+    uniqs = np.unique(a[:, 0]).astype(int)
+
+    inds = np.array([np.where(ids==val)[0][0] for val in uniqs])
+    resids, resnames = ids[inds], names[inds]
+    residues = np.array([f'{name}{resid}' for name, resid in zip(resnames, resids)])
+
+    with open('residue_list.csv', 'w+') as r:
+        for residue in residues:
+            r.write(f'{residue},')
 
-if not os.path.exists('basicrta-7.0'):
-    os.mkdir('basicrta-7.0')
+    if not os.path.exists(f'basicrta-{cutoff}'):
+        os.mkdir(f'basicrta-{cutoff}')
 
-os.chdir('basicrta-7.0')
+    os.chdir(f'basicrta-{cutoff}')
+
+    for residue in residues:
+        if not os.path.exists(residue):
+            os.mkdir(residue)
 
-for residue in residues:
-    if not os.path.exists(residue):
-        os.mkdir(residue)
+

scripts/status_update.sh

@@ -1,4 +1,16 @@
 #!/usr/bin/bash                                                                 
-
+
+if [ -z "$1" ]; then 
+	echo "Provide cutoff"
+	exit 1
+fi
+cutoff=$1
+
 sacct -n -X --format jobname,jobid -s RUNNING > .running_jobs.txt               
-awk -v col1=1 -v col2=2 '{printf "basicrta-7.0/%s/slurm-%s.out \n", $col1, $col2}' < .running_jobs.txt | xargs tail -n 1 | sed 's@[^a-z A-Z0-9.:%|</]@@;s/\].*/\]/'; echo
+filenames=`awk -v col1=1 -v col2=2 '{printf "basicrta-%s/%s/slurm_%s.out \n", "'$cutoff'", $col1, $col2}' < .running_jobs.txt`
+
+for file in $filenames; do
+	if [ -e $file ]; then
+		echo $file | xargs tail -n 1 | sed 's@[^a-z A-Z0-9.:%|</]@@;s/\].*/\]/'; echo
+	fi;
+done

scripts/submit_tmp.slu

@@ -4,8 +4,11 @@
 #SBATCH -N 1
 #SBATCH -n 1
 #SBATCH --mem=20G
-#SBATCH basicrta-7.0/RESIDUE/slurm_${SLURM_JOB_ID}.log
-
-python -m basicrta.gibbs --resid RESID 
+#SBATCH -o basicrta-7.0/RESIDUE/slurm_%j.out
+#SBATCH -t 1-12:00:00
+#SBATCH -p general
+#SBATCH -q public
+#SBATCH --cpus-per-task=1
 
+python -m basicrta.gibbs --contacts contacts_7.0.pkl --resid RESID