1
2
3
4
5
6

# conda install -c conda-forge rdkit

from rdkit import Chem
from rdkit.Chem import MACCSkeys
from rdkit import DataStructs
from rdkit.Chem import Draw

1
2
3
4
5
6
7
8

m = Chem.MolFromSmiles('CCOC')
# Chem.MolToSmiles(mol)
fp = Chem.RDKFingerprint(m, fpSize=1024)
# fpSize 自定义数目,默认为2048

fp.GetNumBits()
# 1024
fp.ToBitString()

1
2
3
4
5
6

from rdkit.Chem import MACCSkeys
fp = MACCSkeys.GenMACCSKeys(m)

fp.GetNumBits()
# 167
fp.ToBitString()

1
2
3
4
5

from rdkit.Chem import AllChem
fp = AllChem.GetMorganFingerprintAsBitVect(m1, radius = 6, nBits = 1024)
fp.GetNumBits()
# 1024
fp.ToBitString()

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27

m = Chem.MolFromSmiles('CCOC')
fp = MACCSkeys.GenMACCSKeys(m)
m2 = Chem.MolFromSmiles('CCO')
fp2 = MACCSkeys.GenMACCSKeys(m2)

DataStructs.FingerprintSimilarity(fp, fp2)
# 0.5

#默认评价相似度指标为Tanimoto
DataStructs.FingerprintSimilarity(fp, fp2, metric=DataStructs.TanimotoSimilarity)
#还有其它指标
metic_list = ['DataStructs.TanimotoSimilarity',
              'DataStructs.DiceSimilarity',
              'DataStructs.CosineSimilarity',
              'DataStructs.SokalSimilarity',
              'DataStructs.RusselSimilarity',
              'DataStructs.KulczynskiSimilarity',
              'DataStructs.McConnaugheySimilarity']
for i in metic_list:
    print(DataStructs.FingerprintSimilarity(fp, fp2, metric=eval(i)))
# 0.5
# 0.6666666666666666
# 0.6735753140545634
# 0.3333333333333333
# 0.041916167664670656
# 0.6805555555555556
# 0.3611111111111111

1
2

imgs = Draw.MolsToGridImage([m,m2],molsPerRow=2,subImgSize=(200,200), legends=["mol-1","mol-2"])
imgs

1
2

pip install padelpy
# pip install pandas

1
2
3
4
5
6
7

unzip fingerprints_xml.zip
mkdir fingerprints_xml
ls *.xml | while read id; do mv $id ./fingerprints_xml/; done
ls ./fingerprints_xml
# AtomPairs2DFingerprintCount.xml  ExtendedFingerprinter.xml   KlekotaRothFingerprintCount.xml  PubchemFingerprinter.xml
# AtomPairs2DFingerprinter.xml     Fingerprinter.xml           KlekotaRothFingerprinter.xml     SubstructureFingerprintCount.xml
# EStateFingerprinter.xml          GraphOnlyFingerprinter.xml  MACCSFingerprinter.xml           SubstructureFingerprinter.xml

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20

import glob
xml_files = glob.glob("./fingerprints_xml/*.xml")
xml_files.sort()
xml_files

FP_list = ['AtomPairs2DCount',
 'AtomPairs2D',
 'EState',
 'CDKextended',
 'CDK',
 'CDKgraphonly',
 'KlekotaRothCount',
 'KlekotaRoth',
 'MACCS',
 'PubChem',
 'SubstructureCount',
 'Substructure']
 
fp = dict(zip(FP_list, xml_files))
fp

1
2
3
4

df = pd.read_csv('./HCV_NS5B_Curated.csv')
df[['CANONICAL_SMILES','CMPD_CHEMBLID']].head()
# 保存文件
df2.head(50).to_csv('molecule.smi', sep='\t', index=False, header=False)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21

from padelpy import padeldescriptor

# 假设需要转换为Substructure指纹
fingerprint = 'Substructure'
fingerprint_output_file = ''.join([fingerprint,'.csv']) #Substructure.csv 结果文件名
fingerprint_descriptortypes = fp[fingerprint] #解析文件地址

padeldescriptor(mol_dir='molecule.smi', 
                d_file=fingerprint_output_file, #'Substructure.csv'
                #descriptortypes='SubstructureFingerprint.xml', 
                descriptortypes= fingerprint_descriptortypes,
                detectaromaticity=True,
                standardizenitro=True,
                standardizetautomers=True,
                threads=2,
                removesalt=True,
                log=True,
                fingerprints=True)

descriptors = pd.read_csv(fingerprint_output_file)
descriptors

1
2
3
4
5
6

df = pd.read_csv('./HCV_NS5B_Curated.csv')
smi = list(df["CANONICAL_SMILES"][0:50])
descriptors = from_smiles(smi)
descriptors_df = pd.DataFrame(descriptors)
descriptors_df.shape
# (50, 1875)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10

# 安装过程，由于版本冲突原因费了一番周折
# (1) python 3.9的环境下
conda install numpy=1.20 networkx=2.3
conda install -c rdkit -c mordred-descriptor mordred

# https://github.com/mordred-descriptor/mordred/issues/84
# (2) 修改DetourMatrix.py模块文件的源代码
# for bcc in networkx.biconnected_component_subgraphs(self.G, False):
# to:
# for bcc in (self.G.subgraph(c) for c in networkx.biconnected_components(self.G)):

1
2
3

# 准备好仅包含一列smiles式的文本文件
python -m mordred tmp/canocical_1_col.smi -p 1 --3D  \
-o ./mordred_descriptor1826.csv