Merge pull request #32 from fwmeng88/master

Version 0.1.11

Add support of SDF file

.gitignore

@@ -0,0 +1,111 @@
+# Prerequisites
+*.d
+
+# Compiled object files
+*.slo
+*.lo
+*.o
+*.obj
+
+# Precompiled headers
+*.gch
+*.pch
+
+# Compiled dynamic libraries
+*.so
+*.so.[0-9]*
+*.dylib
+*.dll
+
+# Fortran module files
+*.mod
+*.smod
+
+# Compiled static libraries
+*.lai
+*.la
+*.a
+*.lib
+
+# Executables
+*.exe
+*.out
+*.app
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+.installed.cfg
+MANIFEST
+*.egg-info/
+*.egg
+*.manifest
+*.spec
+pip-log.txt
+pip-delete-this-directory.txt
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*,cover
+.pytest_cache/
+
+# Documentation
+doc/html/
+doc/latex/
+doc/man/
+doc/xml/
+doc/_build/
+doc/source
+doc/modules
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+
+# Editor junk
+tags
+[._]*.s[a-v][a-z]
+[._]*.sw[a-p]
+[._]s[a-v][a-z]
+[._]sw[a-p]
+*~
+\#*\#
+.\#*
+.ropeproject
+.idea/
+.spyderproject
+.spyproject
+.vscode/
+# Mac .DS_Store
+.DS_Store
+
+# jupyter notebook checkpoints
+.ipynb_checkpoints
+
+# version file generated by rob
+padelpy/_version.py
+

README.md

@@ -74,6 +74,28 @@ fingerprints = from_mdl('mols.mdl', fingerprints=True, descriptors=False)
 _ = from_mdl('mols.mdl', output_csv='descriptors.csv')
 ```
 
+### SDF to Descriptors/Fingerprints
+
+The "from_sdf" function accepts a filepath as an argument, and returns a list.
+Each list element is a dictionary with descriptors/fingerprints corresponding to each supplied
+molecule (indexed as they appear in the SDF file).
+
+```python
+from padelpy import from_sdf
+
+# calculate molecular descriptors for molecules in `mols.sdf`
+descriptors = from_sdf('mols.sdf')
+
+# in addition to descriptors, calculate PubChem fingerprints
+desc_fp = from_sdf('mols.sdf', fingerprints=True)
+
+# only calculate fingerprints
+fingerprints = from_sdf('mols.sdf', fingerprints=True, descriptors=False)
+
+# save descriptors to a CSV file
+_ = from_sdf('mols.sdf', output_csv='descriptors.csv')
+```
+
 ### Command Line Wrapper
 
 Alternatively, you can have more control over PaDEL-Descriptor with the command-line wrapper function. Any combination of arguments supported by PaDEL-Descriptor can be accepted by the "padeldescriptor" function.
@@ -84,9 +106,12 @@ from padelpy import padeldescriptor
 # to supply a configuration file
 padeldescriptor(config='\\path\\to\\config')
 
-# to supply an input and output file
+# to supply an input (MDL) and output file
 padeldescriptor(mol_dir='molecules.mdl', d_file='descriptors.csv')
 
+# to supply an input (SDF) and output file
+padeldescriptor(mol_dir='molecules.sdf', d_file='descriptors.csv')
+
 # a SMILES file can be supplied
 padeldescriptor(mol_dir='molecules.smi', d_file='descriptors.csv')
 

padelpy/__init__.py

@@ -1,3 +1,3 @@
-from padelpy.wrapper import padeldescriptor
-from padelpy.functions import from_mdl, from_smiles
+from .wrapper import padeldescriptor
+from .functions import from_mdl, from_smiles, from_sdf
 __version__ = '0.1.10'

padelpy/functions.py

@@ -8,22 +8,28 @@
 # Contains various functions commonly used with PaDEL-Descriptor
 #
 
+import warnings
 # stdlib. imports
 from collections import OrderedDict
 from csv import DictReader
 from datetime import datetime
 from os import remove
 from re import compile, IGNORECASE
 from time import sleep
-import warnings
 
 # PaDELPy imports
 from padelpy import padeldescriptor
 
+__all__ = [
+    "from_mdl",
+    "from_smiles",
+    "from_sdf",
+]
+
 
 def from_smiles(smiles, output_csv: str = None, descriptors: bool = True,
                 fingerprints: bool = False, timeout: int = 60) -> OrderedDict:
-    ''' from_smiles: converts SMILES string to QSPR descriptors/fingerprints
+    """ from_smiles: converts SMILES string to QSPR descriptors/fingerprints
 
     Args:
         smiles (str, list): SMILES string for a given molecule, or a list of
@@ -38,30 +44,30 @@ def from_smiles(smiles, output_csv: str = None, descriptors: bool = True,
             list of OrderedDicts, else single OrderedDict; each OrderedDict
             contains labels and values for each descriptor generated for each
             supplied molecule
-    '''
+    """
 
-    timestamp = datetime.now().strftime('%Y%m%d%H%M%S%f')[:-3]
+    timestamp = datetime.now().strftime("%Y%m%d%H%M%S%f")[:-3]
 
-    with open('{}.smi'.format(timestamp), 'w') as smi_file:
+    with open("{}.smi".format(timestamp), "w") as smi_file:
         if type(smiles) == str:
             smi_file.write(smiles)
         elif type(smiles) == list:
-            smi_file.write('\n'.join(smiles))
+            smi_file.write("\n".join(smiles))
         else:
-            raise RuntimeError('Unknown input format for `smiles`: {}'.format(
+            raise RuntimeError("Unknown input format for `smiles`: {}".format(
                 type(smiles)
             ))
     smi_file.close()
 
     save_csv = True
     if output_csv is None:
         save_csv = False
-        output_csv = '{}.csv'.format(timestamp)
+        output_csv = "{}.csv".format(timestamp)
 
     for attempt in range(3):
         try:
             padeldescriptor(
-                mol_dir='{}.smi'.format(timestamp),
+                mol_dir="{}.smi".format(timestamp),
                 d_file=output_csv,
                 convert3d=True,
                 retain3d=True,
@@ -74,7 +80,7 @@ def from_smiles(smiles, output_csv: str = None, descriptors: bool = True,
             break
         except RuntimeError as exception:
             if attempt == 2:
-                remove('{}.smi'.format(timestamp))
+                remove("{}.smi".format(timestamp))
                 if not save_csv:
                     sleep(0.5)
                     try:
@@ -85,33 +91,33 @@ def from_smiles(smiles, output_csv: str = None, descriptors: bool = True,
             else:
                 continue
 
-    with open(output_csv, 'r', encoding='utf-8') as desc_file:
+    with open(output_csv, "r", encoding="utf-8") as desc_file:
         reader = DictReader(desc_file)
         rows = [row for row in reader]
     desc_file.close()
 
-    remove('{}.smi'.format(timestamp))
+    remove("{}.smi".format(timestamp))
     if not save_csv:
         remove(output_csv)
 
     if type(smiles) == list and len(rows) != len(smiles):
-        raise RuntimeError('PaDEL-Descriptor failed on one or more mols.' +
-                           ' Ensure the input structures are correct.')
+        raise RuntimeError("PaDEL-Descriptor failed on one or more mols." +
+                           " Ensure the input structures are correct.")
     elif type(smiles) == str and len(rows) == 0:
         raise RuntimeError(
-            'PaDEL-Descriptor failed on {}.'.format(smiles) +
-            ' Ensure input structure is correct.'
+            "PaDEL-Descriptor failed on {}.".format(smiles) +
+            " Ensure input structure is correct."
         )
 
     for idx, r in enumerate(rows):
         if len(r) == 0:
             raise RuntimeError(
-                'PaDEL-Descriptor failed on {}.'.format(smiles[idx]) +
-                ' Ensure input structure is correct.'
+                "PaDEL-Descriptor failed on {}.".format(smiles[idx]) +
+                " Ensure input structure is correct."
             )
 
     for idx in range(len(rows)):
-        del rows[idx]['Name']
+        del rows[idx]["Name"]
 
     if type(smiles) == str:
         return rows[0]
@@ -120,7 +126,7 @@ def from_smiles(smiles, output_csv: str = None, descriptors: bool = True,
 
 def from_mdl(mdl_file: str, output_csv: str = None, descriptors: bool = True,
              fingerprints: bool = False, timeout: int = 60) -> list:
-    ''' from_mdl: converts MDL file into QSPR descriptors/fingerprints;
+    """ from_mdl: converts MDL file into QSPR descriptors/fingerprints;
     multiple molecules may be represented in the MDL file
 
     Args:
@@ -133,25 +139,69 @@ def from_mdl(mdl_file: str, output_csv: str = None, descriptors: bool = True,
     Returns:
         list: list of dicts, where each dict corresponds sequentially to a
             compound in the supplied MDL file
-    '''
+    """
 
-    is_mdl = compile(r'.*\.mdl$', IGNORECASE)
+    is_mdl = compile(r".*\.mdl$", IGNORECASE)
     if is_mdl.match(mdl_file) is None:
-        raise ValueError('MDL file must have a `.mdl` extension: {}'.format(
+        raise ValueError("MDL file must have a `.mdl` extension: {}".format(
             mdl_file
         ))
 
+    rows = _from_mdl_lower(mol_file=mdl_file,
+                           output_csv=output_csv,
+                           descriptors=descriptors,
+                           fingerprints=fingerprints,
+                           timeout=timeout)
+    return rows
+
+
+def from_sdf(sdf_file: str,
+             output_csv: str = None,
+             descriptors: bool = True,
+             fingerprints: bool = False,
+             timeout: int = 60) -> list:
+    """ Converts sdf file into QSPR descriptors/fingerprints.
+    Multiple molecules may be represented in the sdf file
+
+    Args:
+        sdf_file (str): path to sdf file
+        output_csv (str): if supplied, saves descriptors/fingerprints here
+        descriptors (bool): if `True`, calculates descriptors
+        fingerprints (bool): if `True`, calculates fingerprints
+        timeout (int): maximum time, in seconds, for conversion
+
+    Returns:
+        list: list of dicts, where each dict corresponds sequentially to a compound in the
+        supplied sdf file
+    """
+
+    is_sdf = compile(r".*\.sdf$", IGNORECASE)
+    if is_sdf.match(sdf_file) is None:
+        raise ValueError("sdf file must have a `.sdf` extension: {}".format(
+            sdf_file
+        ))
+
+    rows = _from_mdl_lower(mol_file=sdf_file,
+                           output_csv=output_csv,
+                           descriptors=descriptors,
+                           fingerprints=fingerprints,
+                           timeout=timeout)
+    return rows
+
+
+def _from_mdl_lower(mol_file: str, output_csv: str = None, descriptors: bool = True,
+                    fingerprints: bool = False, timeout: int = 60) -> list:
     save_csv = True
     if output_csv is None:
         save_csv = False
-        output_csv = '{}.csv'.format(
-            datetime.now().strftime('%Y%m%d%H%M%S%f')[:-3]
+        output_csv = "{}.csv".format(
+            datetime.now().strftime("%Y%m%d%H%M%S%f")[:-3]
         )
 
     for attempt in range(3):
         try:
             padeldescriptor(
-                mol_dir=mdl_file,
+                mol_dir=mol_file,
                 d_file=output_csv,
                 convert3d=True,
                 retain3d=True,
@@ -174,15 +224,16 @@ def from_mdl(mdl_file: str, output_csv: str = None, descriptors: bool = True,
             else:
                 continue
 
-    with open(output_csv, 'r', encoding='utf-8') as desc_file:
+    with open(output_csv, "r", encoding="utf-8") as desc_file:
         reader = DictReader(desc_file)
         rows = [row for row in reader]
     desc_file.close()
     if not save_csv:
         remove(output_csv)
     if len(rows) == 0:
-        raise RuntimeError('PaDEL-Descriptor returned no calculated values.' +
-                           ' Ensure the input structure is correct.')
+        raise RuntimeError("PaDEL-Descriptor returned no calculated values." +
+                           " Ensure the input structure is correct.")
     for row in rows:
-        del row['Name']
+        del row["Name"]
+
     return rows

padelpy/version.py

@@ -0,0 +1,5 @@
+
+VERSION = (0, 1, 11, "")
+
+__version__ = ".".join(map(str, VERSION[:-1]))
+__release__ = ".".join(map(str, VERSION))