Update code/alphafold_featureVector.py

code/alphafold_featureVector.py

@@ -44,6 +44,53 @@ import requests
 from Bio.PDB import PDBParser, PPBuilder
 from io import StringIO
 
+import glob
+import ssbio.utils
+import subprocess
+import ssbio
+import os.path as op
+from add_3Dalignment import *
+import os
+from pathlib import Path
+import gzip
+import shutil
+import streamlit as st
+
+
+def run_freesasa(infile, outfile, include_hetatms=True, outdir=None, force_rerun=False, file_type = 'gzip'):
+    if not outdir:
+        outdir = ''
+    outfile = op.join(outdir, outfile)
+    if file_type == 'pdb':
+        if ssbio.utils.force_rerun(flag=force_rerun, outfile=outfile):
+            if include_hetatms:
+                shell_command = 'freesasa --format=rsa --hetatm {} -o {}'.format(infile, outfile)
+            else:
+                shell_command = 'freesasa --format=rsa {} -o {}'.format(infile, outfile)
+            command = subprocess.Popen(shell_command,
+                                       stdout=subprocess.PIPE,
+                                       stderr=subprocess.PIPE,
+                                       shell=True)
+            out, err = command.communicate()
+    elif file_type == 'gzip':
+        with gzip.open(infile, 'rb') as f_in:
+            with open('file_temp.pdb', 'wb') as f_out:
+                shutil.copyfileobj(f_in, f_out)
+
+        infile = 'file_temp.pdb'
+
+        if ssbio.utils.force_rerun(flag=force_rerun, outfile=outfile):
+            if include_hetatms:
+                shell_command = 'freesasa --format=rsa --hetatm {} -o {}'.format(infile, outfile)
+            else:
+                shell_command = 'freesasa --format=rsa {} -o {}'.format(infile, outfile)
+            command = subprocess.Popen(shell_command,
+                                       stdout=subprocess.PIPE,
+                                       stderr=subprocess.PIPE,
+                                       shell=True)
+            out, err = command.communicate()
+    return outfile
+
 
 def convert_non_standard_amino_acids(sequence):
     """
@@ -629,7 +676,6 @@ def alphafold(input_set, mode, impute):
                 atomSequence = convert_non_standard_amino_acids(atomSequence)
                 alignments = aligner.align(pdbSequence, atomSequence)
                 alignments = (list(alignments))
-                            
                 #if get_alignments_3D(uniprotID, mod, pdb_path, pdbSequence, 'nan', 'nan', 'nan', mode, Path(path_to_output_files / '3D_alignment'),
                 #                     'gzip') != None:
 
@@ -640,8 +686,13 @@ def alphafold(input_set, mode, impute):
 
                     
                     alignments = alignments[0]
-
-                    calculate_freesasa(uniprotID, mod, existing_free_sasa, alphafold_path, path_to_output_files)
+                    if ID not in existing_free_sasa:
+                        fullID = f'AF-{ID}-F{model_num}-{file_str }.pdb.gz'
+                        run_freesasa(Path(alphafold_path / fullID),
+                                     Path(path_to_output_files / f'freesasa_files/{fullID}.txt'), include_hetatms=True,
+                                     outdir=None, force_rerun=False)
+                    st.write('Calulated')
+                    #calculate_freesasa(uniprotID, mod, existing_free_sasa, alphafold_path, path_to_output_files)
                     if (mutationPositionOnPDB != 'nan'):
                         if (int(mutationPositionOnPDB) <= 1400):
                             try: