PaDELPy provides a Python wrapper for the PaDEL-Descriptor molecular descriptor calculation software. It was created to allow direct access to the PaDEL-Descriptor command-line interface via Python.
Installation via pip:
$ pip install padelpy
Installation via cloned repository:
$ git clone https://github.com/ecrl/padelpy
$ cd padelpy
$ pip install .
PaDEL-Descriptor is bundled into PaDELPy, therefore an external installation/download of PaDEL-Descriptor is not necessary. There are currently no additional Python dependencies for PaDELPy, however it requires an installation of the Java JRE version 6+.
In addition to providing a complete interface between Python and PaDEL-Descriptor's command line tool, PaDELPy offers two functions to acquire descriptors/fingerprints within Python - obtaining descriptors/fingerprints from a SMILES string, and obtaining descriptors/fingerprints from an MDL MolFile.
The "from_smiles" function accepts a SMILES string or list of SMILES strings as an argument, and returns a Python dictionary with descriptor/fingerprint names/values as keys/values respectively - if multiple SMILES strings are supplied, "from_smiles" returns a list of dictionaries.
from padelpy import from_smiles
# calculate molecular descriptors for propane
descriptors = from_smiles('CCC')
# calculate molecular descriptors for propane and butane
descriptors = from_smiles(['CCC', 'CCCC'])
# in addition to descriptors, calculate PubChem fingerprints
desc_fp = from_smiles('CCC', fingerprints=True)
# only calculate fingerprints
fingerprints = from_smiles('CCC', fingerprints=True, descriptors=False)
# setting the number of threads, this uses one cpu thread to compute descriptors
descriptors = from_smiles(['CCC', 'CCCC'], threads = 1)
# save descriptors to a CSV file
_ = from_smiles('CCC', output_csv='descriptors.csv')The "from_mdl" function accepts a filepath (to an MDL MolFile) 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 MolFile).
from padelpy import from_mdl
# calculate molecular descriptors for molecules in `mols.mdl`
descriptors = from_mdl('mols.mdl')
# in addition to descriptors, calculate PubChem fingerprints
desc_fp = from_mdl('mols.mdl', fingerprints=True)
# only calculate fingerprints
fingerprints = from_mdl('mols.mdl', fingerprints=True, descriptors=False)
# setting the number of threads, this uses one cpu thread to compute descriptors
desc_fp = from_mdl('mols.mdl', threads=1)
# save descriptors to a CSV file
_ = from_mdl('mols.mdl', output_csv='descriptors.csv')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).
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)
# setting the number of threads, this uses one cpu thread to compute descriptors
desc_fp = from_mdl('mols.sdf', threads=1)
# save descriptors to a CSV file
_ = from_sdf('mols.sdf', output_csv='descriptors.csv')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.
from padelpy import padeldescriptor
# to supply a configuration file
padeldescriptor(config='\\path\\to\\config')
# 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')
# a path to a directory containing structural files can be supplied
padeldescriptor(mol_dir='\\path\\to\\mols\\', d_file='descriptors.csv')
# to calculate 2-D and 3-D descriptors
padeldescriptor(d_2d=True, d_3d=True)
# to calculate PubChem fingerprints
padeldescriptor(fingerprints=True)
# to convert molecule into a 3-D structure
padeldescriptor(convert3d=True)
# to supply a descriptortypes file
padeldescriptor(descriptortype='\\path\\to\\descriptortypes')
# to detect aromaticity
padeldescriptor(detectaromaticity=True)
# to calculate fingerprints
padeldescriptor(fingerprints=True)
# to save process status to a log file
padeldescriptor(log=True)
# to remove salts from the molecule(s)
padeldescriptor(removesalt=True)
# to retain 3-D coordinates when standardizing
padeldescriptor(retain3d=True)
# to retain order (output same order as input)
padeldescriptor(retainorder=True)
# to standardize nitro groups to N(:O):O
padeldescriptor(standardizenitro=True)
# to standardize tautomers
padeldescriptor(standardizetautomers=True)
# to specify a SMIRKS tautomers file
padeldescriptor(tautomerlist='\\path\\to\\tautomers\\')
# to use filenames as molecule names
padeldescriptor(usefilenameasmolname=True)
# to set the maximum number of compounds in a resulting descriptors file
padeldescriptor(maxcpdperfile=32)
# to set the maximum runtime (in mS) per molecule
padeldescriptor(maxruntime=10000)
# to set the maximum number of waiting jobs in the queue
padeldescriptor(waitingjobs=10)
# to set the maximum number of threads used
padeldescriptor(threads=2)
# to prevent padel-splash image from loading.
padeldescriptor(headless=True)To contribute to PaDELPy, make a pull request. Contributions should include tests for new features added, as well as extensive documentation.
To report problems with the software or feature requests, file an issue. When reporting problems, include information such as error messages, your OS/environment and Python version.
For additional support/questions, contact Travis Kessler ([email protected]).