rdkit Recap、BRICS将smiles切分为片段（fragment）

一、 Recap切分

主体函数 

from rdkit import Chem
from rdkit.Chem import BRICS
import numpy as np
 
 
# --------------------- 将分子拆分为不带数字或者自定义 ---------------------
def fragment_recursive(mol, frags):
    try:
        bonds = list(BRICS.FindBRICSBonds(mol))
        if len(bonds) == 0:
            frags.append(Chem.MolToSmiles(mol))
            return frags
 
        idxs, labs = list(zip(*bonds))
        bond_idxs = []
        for a1, a2 in idxs:
            bond = mol.GetBondBetweenAtoms(a1, a2)
            bond_idxs.append(bond.GetIdx())
        order = np.argsort(bond_idxs).tolist()
        bond_idxs = [bond_idxs[i] for i in order]
        broken = Chem.FragmentOnBonds(mol, bondIndices=[bond_idxs[0]], dummyLabels=[(0, 0)])
        head, tail = Chem.GetMolFrags(broken, asMols=True)
        frags.append(Chem.MolToSmiles(head))
        return fragment_recursive(tail, frags)
    except Exception as e:
        print(e)
        pass
 
 
# --------------------- 将*号去掉 ---------------------
def remove_dummy(smiles):
    try:
        stripped_smi=smiles.replace('*','[H]')
        mol=Chem.MolFromSmiles(stripped_smi)
        return Chem.MolToSmiles(mol)
    except Exception as e:
        print(e)
        return None

1、单个smiles拆分为fragment

# 单个smiles拆分为fragment
aspirin = Chem.MolFromSmiles('C1CC1C(=O)N2CCN(CC2)C(=O)C3=C(C=CC(=C3)CC4=NNC(=O)C5=CC=CC=C54)F')
fragments = fragment_recursive(aspirin, [])
clean_fragments = [remove_dummy(smi) for smi in fragments]
print(clean_fragments)

2、csv文件的每一行smiles拆分为fragment + 删除重复的行 + 存入新文件中

# csv文件的每一行smiles拆分为fragment + 删除重复的行
import csv
import pandas as pd
 
# 读取原始 CSV 文件
with open('csv/double.csv', 'r') as csv_file:
    reader = csv.reader(csv_file)
    input_data = list(reader)
 
# 处理每一行数据并将结果保存到列表中
output_data = []
for row in input_data:
    try:
        aspirin = Chem.MolFromSmiles(row[0])
        fragments = fragment_recursive(aspirin, [])
        clean_fragments = [remove_dummy(smi) for smi in fragments]
        output_data.append(clean_fragments)
    except:
        print("Current smiles process Error: ", row)
 
# 存入新的csv文件 + delete duplicate item
new_csv_name = 'output.smi'
df = pd.DataFrame(output_data)
df = df.stack().reset_index(drop=True)
df.to_csv(new_csv_name, index=False, header=False)
df = pd.read_csv(new_csv_name, header=None)
df.drop_duplicates(inplace=True)
df.to_csv(new_csv_name, index=False, header=False)

二、 BRICS切分

rdkit的BRICS算法，BRICS基于常见的反应，选择片段断键的位点，提供了化学合成意义上的可行性。

from rdkit.Chem import Recap
from rdkit.Chem import AllChem as Chem
 
m = Chem.MolFromSmiles('c1ccccc1OCCOC(=O)CC')
hierarch = Recap.RecapDecompose(m)
 
 
#叶子节点函数：hierarch.GetLeaves()
print(hierarch.GetLeaves().keys())
 
#子孙节点函数：hierarch.GetAllChildren()
print(hierarch.GetAllChildren().keys())
 
#祖先节点函数，返回列表：getUltimateParents()
print(hierarch.getUltimateParents()[0].smiles)
 

药物分子进行片段分解 - 知乎

原文来源于：rdkit Recap、BRICS分子片段拆分与合成_loong_XL的博客-CSDN博客