Update code/pdb_featureVector.py

code/pdb_featureVector.py

@@ -286,6 +286,7 @@ def pdb(input_set, mode, impute):
 
         uniprot_matched = uniprot_matched.merge(pdb_fasta, on=['pdbID', 'chain'], how='left')
         uniprot_matched = uniprot_matched.astype(str)
+        st.write('uniprot_matched', uniprot_matched)
 
         with_pdb = uniprot_matched[(uniprot_matched.pdbID != 'nan') & (
                 (uniprot_matched.resolution != 'nan') & (uniprot_matched.resolution != 'OT') & (
@@ -295,11 +296,12 @@ def pdb(input_set, mode, impute):
                 uniprot_matched.resolution == 'None'))]
         no_pdb = no_pdb[~no_pdb.datapoint.isin(with_pdb.datapoint.to_list())]
         no_pdb.drop(columns=['chain', 'pdbID', 'pdbSequence', 'resolution'], inplace=True)
-
+        st.write('with_pdb', with_pdb)
         print(
             'PDB Information successfully added...\nPDB structures are found for %d of %d.\n%d of %d failed to match with PDB structure.\n'
             % (len(with_pdb.drop_duplicates(['datapoint'])), len(uniprot_matched.drop_duplicates(['datapoint'])),
                len(no_pdb.drop_duplicates(['datapoint'])), len(uniprot_matched.drop_duplicates(['datapoint']))))
+        st.write('with_pdb1', with_pdb)
 
         with_pdb = with_pdb.sort_values(['uniprotID', 'resolution'], axis=0, ascending=True)
         with_pdb = with_pdb.drop_duplicates(['uniprotID', 'wt', 'mut', 'pos', 'pdbSequence'], keep='first')
@@ -378,6 +380,7 @@ def pdb(input_set, mode, impute):
         # Isoform matches, i.e. labelled as i, isoform sequences will be aligned with PDB sequences.
         with_pdb['uniprotSequence'] = with_pdb['uniprotSequence'].str.replace('U', 'C')
         with_pdb['pdbSequence'] = with_pdb['pdbSequence'].str.replace('U', 'C')
+        st.write('with_pdb2', with_pdb)
 
         dfM = with_pdb[with_pdb.wt_sequence_match == 'm']
         dfM = dfM.sort_values(['uniprotID', 'resolution'], axis=0, ascending=True)