Introduction

Installation

conda(recommended)

1. install conda

   • miniconda
   • anaconda

2. install mordred

   stable

   $ conda install -c rdkit -c mordred-descriptor mordred
   

   beta

   $ conda install -c rdkit -c mordred-descriptor/channel/dev mordred
   

pip

1. install rdkit python package

2. install mordred

   $ pip install git+https://github.com/mordred-descriptor/mordred
   

examples

as command

all descriptors

$ python -m mordred --help
usage: python -m mordred [-h] [-i PATH] [-f TYPE] [-o [PATH]] [-p N]

optional arguments:
  -h, --help            show this help message and exit
  -i PATH, --input PATH
                        input file or directory(default: stdin)
  -f TYPE, --from TYPE  input filetype(one of auto, smi, sdf, mol, default:
                        auto)
  -o [PATH], --output [PATH]
                        output csv file(default: stdout)
  -p N, --processes N   number of processes to use(default: number of threads)

descriptors in submodule

$ python -m mordred.TPSA -i tests/references/structures.smi
name,TPSA(NO),TPSA
Hexane,0.0,0.0
Benzene,0.0,0.0
Caffeine,61.82,61.82
Cyanidin,112.45,112.45
Lycopene,0.0,0.0
Epicatechin,110.38,110.38
Limonene,0.0,0.0
Allicin,17.07,61.58
Glutathione,158.82,197.62

as library

from rdkit import Chem

from mordred import Calculator, all_descriptors

# create descriptor calculator with all descriptors
calc = Calculator(all_descriptors())

# calculate and print descriptors
for desc, value in calc(Chem.MolFromSmiles('c1ccccc1O')):
   print('{}\t{}'.format(desc, value))