statistical potentials

+# before executing this script, you have to download the RNA-puzzles 3D structures files
+# this script is used to transform the 3D RNA-puzzle structure files from PDB format to mmCIF format 
+# after that, we can perform the geometric measurements on these files by changing the file paths in geometric_stats.py
+
+import Bio
+from Bio.PDB import PDBParser
+from Bio.PDB import MMCIFParser
+import os
+from os import path
+import warnings
+import numpy as np
+import pandas as pd
+from Bio.PDB.vectors import Vector, calc_angle, calc_dihedral
+from statistics import get_euclidian_distance, get_flat_angle, get_torsion_angle, pos_b1, pos_b2
+from multiprocessing import Pool
+
+runDir = os.getcwd()
+
+os.makedirs(runDir+"/results_decoy/geometry/HiRE-RNA/distances/", exist_ok=True)
+os.makedirs(runDir+"/results_decoy/geometry/HiRE-RNA/angles/", exist_ok=True)
+os.makedirs(runDir+"/results_decoy/geometry/HiRE-RNA/torsions/", exist_ok=True)
+
+def split_res_name(f):
+    name=f.split('.')[0]
+
+    if path.isfile(runDir+"/RNA-puzzles/rna_predict/"+name+".cif"):
+        return
+    
+    with warnings.catch_warnings():
+        # Ignore the PDB problems. This mostly warns that some chain is discontinuous.
+        warnings.simplefilter('ignore', Bio.PDB.PDBExceptions.PDBConstructionWarning)
+        warnings.simplefilter('ignore', Bio.PDB.PDBExceptions.BiopythonWarning)
+        parser=PDBParser()
+        s = parser.get_structure(name, os.path.abspath(runDir + "/RNA-puzzles/raw_dataset_and_for_assessment/rna_predict/" + f ))
+    new_s=Bio.PDB.Structure.Structure(s.get_id())
+    model=s[0]
+    new_model=Bio.PDB.Model.Model(model.get_id())
+    for chain in model:
+        new_chain=Bio.PDB.Chain.Chain(chain.get_id())
+        for res in chain:
+            res_atoms=res.get_atoms()
+            new_residu=Bio.PDB.Residue.Residue(res.get_id(), res.get_resname().split('  ')[1], res.get_segid())
+            for atom in list(res.get_atoms()):
+                new_atom=atom.copy()
+                new_residu.add(new_atom)
+            new_chain.add(new_residu)
+
+        new_model.add(new_chain)
+    new_s.add(new_model)
+            
+            
+    ioobj = Bio.PDB.MMCIFIO()
+    ioobj.set_structure(new_s)
+    ioobj.save(os.path.abspath(runDir +"/RNA-puzzles/rna_predict/"+name+".cif"))
+
+liste=os.listdir(runDir + '/RNA-puzzles/raw_dataset_and_for_assessment/rna_predict')
+
+
+def main():
+    with Pool(processes=50) as pool:
+        result=pool.map(split_res_name, [f for f in liste])
+
+if __name__ == '__main__':
+    main()
+

+# This script is used to run DSSR on the 3D structures of RNA-puzzles,
+# and then to extract the values of torsion angles and pseudotorsions
+
+
+import json
+import os
+import subprocess
+import warnings
+import pandas as pd
+from multiprocessing import Pool 
+
+runDir = os.getcwd()
+
+os.makedirs(runDir + "/RNA-puzzles/annotations/", exist_ok=True)
+def annotate(fichier):
+    name=fichier.split('/')[-1]
+    puz_id=name.split('.')[0]
+    
+    if (os.path.isfile(runDir +"/RNA-puzzles/annotations/" + puz_id + ".json")):
+        return
+
+    if name.split('.')[1]=='cif':
+    # run DSSR (you need to have it in your $PATH, follow x3dna installation instructions)
+        output = subprocess.run(["x3dna-dssr", f"-i={fichier}", "--json", "--auxfile=no", f"-o={puz_id}.json"], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        stdout = output.stdout.decode('utf-8')
+        stderr = output.stderr.decode('utf-8')
+        
+        if "exception" in stderr:
+            # DSSR is unable to parse the chain.
+            warn(f"Exception while running DSSR, ignoring {puz_id}.", error=True)
+            return 1
+    else:
+        return
+    # save the analysis to file only if we can load it :/
+    
+    json_file = open(runDir + "/RNA-puzzles/annotations/" +
+                 puz_id + ".json", "w")
+    json_file.write(stdout)
+    json_file.close()
+    
+
+
+for f in os.listdir(runDir +"/RNA-puzzles/rna_predict"):
+    annotate(runDir +"/RNA-puzzles/rna_predict/"+ f)
+
+os.makedirs(runDir + "/RNA-puzzles/torsions/", exist_ok=True)
+os.makedirs(runDir +"/RNA-puzzles/pseudotorsions/", exist_ok=True)
+
+def extract_3d_data(f):
+    pdb_id=f.split('.')[0]
+    if f.split('.')[1]=='json':
+        try :
+            with open(runDir + "/RNA-puzzles/rna_predict/" + pdb_id + ".json", 'r') as json_file:
+                json_object = json.load(json_file)
+
+        except json.decoder.JSONDecodeError as e:
+            #warn("Could not load "+pdb_id+f".json with JSON package: {e}", error=True)
+            return None
+
+        # Create the Pandas DataFrame for the nucleotides of the right chain
+        nts = json_object["nts"]                        # sub-json-object
+        df = pd.DataFrame(nts)                       
+        #print(df)
+        cols_we_keep = ["index_chain", "nt_resnum", "nt_name", "nt_code", "nt_id", "dbn", "alpha", "beta", "gamma", "delta", "epsilon", "zeta", "epsilon_zeta", "bb_type", "chi", "glyco_bond", "form", "ssZp", "Dp", "eta", "theta", "eta_prime", "theta_prime", "eta_base", "theta_base", "v0", "v1", "v2", "v3", "v4", "amplitude", "phase_angle", "puckering"]
+        df = df[cols_we_keep]
+        torsions=df[["index_chain", "alpha", "beta", "gamma", "delta", "epsilon", "zeta", "chi"]]
+        pseudotorsions=df[["eta", "theta", "eta_prime", "theta_prime", "eta_base", "theta_base"]]
+        torsions.to_csv('/home/atabot/RNANet/RNA-puzzles/torsions/torsions ' + pdb_id + '.csv')
+        pseudotorsions.to_csv('/home/atabot/RNANet/RNA-puzzles/pseudotorsions/pseudotorsions ' + pdb_id + '.csv')
+        
+    else:
+        return
+    
+
+
+l=os.listdir(runDir + "/RNA-puzzles/rna_predict/")
+
+def main():
+    with Pool(processes=30) as pool:
+        result=pool.map(extract_3d_data, [f for f in l])
+
+if __name__ == '__main__':
+    main()