96 lines
3.5 KiB
Python
96 lines
3.5 KiB
Python
"""
|
||
-------------------------------------------------------------
|
||
批量2D SDF文件 → 3D SDF文件 并行转换脚本 (joblib版)
|
||
-------------------------------------------------------------
|
||
功能：
|
||
- 从指定目录读取2D SDF文件（每个分子一个SDF文件），并行批量生成3D SDF文件，保留原有属性
|
||
- 支持失败分子详细记录和统计
|
||
- 源文件与目标3D SDF目录分离，防止覆盖
|
||
|
||
依赖安装：
|
||
conda install -y -c conda-forge rdkit joblib
|
||
|
||
用法示例：
|
||
python sdf2to3d.py --src_dir ./2d_sdf_dir --out_dir ./3d_sdf_dir --n_jobs 8
|
||
|
||
作者：chatgpt
|
||
-------------------------------------------------------------
|
||
"""
|
||
|
||
from pathlib import Path
|
||
from rdkit import Chem
|
||
from rdkit.Chem import AllChem
|
||
from joblib import Parallel, delayed
|
||
import argparse
|
||
|
||
def convert_2d_to_3d_sdf(sdf_path, out_dir, max_attempts=10):
|
||
identifier = sdf_path.stem
|
||
out_sdf = out_dir / f"{identifier}.sdf"
|
||
try:
|
||
suppl = Chem.SDMolSupplier(str(sdf_path), sanitize=True)
|
||
mols = [mol for mol in suppl if mol is not None]
|
||
if not mols:
|
||
return identifier, False, "SDF读取失败"
|
||
mol = mols[0]
|
||
mol = Chem.AddHs(mol)
|
||
params = AllChem.ETKDGv3()
|
||
last_error = ""
|
||
for attempt in range(max_attempts):
|
||
try:
|
||
status = AllChem.EmbedMolecule(mol, params)
|
||
if status == 0:
|
||
AllChem.UFFOptimizeMolecule(mol)
|
||
writer = Chem.SDWriter(str(out_sdf))
|
||
writer.write(mol)
|
||
writer.close()
|
||
return identifier, True, f"成功（第{attempt+1}次）"
|
||
else:
|
||
last_error = f"Embed失败: status={status}"
|
||
except Exception as e:
|
||
last_error = f"3D生成异常: {e}"
|
||
return identifier, False, last_error if last_error else f"3D构象生成失败（已重试{max_attempts}次）"
|
||
except Exception as e:
|
||
return identifier, False, f"异常: {e}"
|
||
|
||
def main():
|
||
parser = argparse.ArgumentParser()
|
||
parser.add_argument('--src_dir', type=str, required=True, help='2D SDF文件夹路径')
|
||
parser.add_argument('--out_dir', type=str, required=True, help='3D SDF输出文件夹路径')
|
||
parser.add_argument('--n_jobs', type=int, default=4, help='并行进程数')
|
||
parser.add_argument('--max_attempts', type=int, default=10, help='最大Embed尝试次数')
|
||
args = parser.parse_args()
|
||
|
||
src_dir = Path(args.src_dir)
|
||
out_dir = Path(args.out_dir)
|
||
out_dir.mkdir(parents=True, exist_ok=True)
|
||
|
||
sdf_files = list(src_dir.glob("*.sdf"))
|
||
print(f"共检测到2D SDF文件 {len(sdf_files)} 个，开始并行3D生成...")
|
||
|
||
results = Parallel(n_jobs=args.n_jobs)(
|
||
delayed(convert_2d_to_3d_sdf)(sdf_file, out_dir, args.max_attempts) for sdf_file in sdf_files
|
||
)
|
||
|
||
# 分类统计
|
||
success = [r for r in results if r[1]]
|
||
failed = [r for r in results if not r[1]]
|
||
|
||
print("="*60)
|
||
print(f"总2D SDF: {len(sdf_files)}")
|
||
print(f"成功生成3D SDF: {len(success)}")
|
||
print(f"失败: {len(failed)}")
|
||
if failed:
|
||
print("失败分子列表：")
|
||
for ident, _, msg in failed:
|
||
print(f" {ident}: {msg}")
|
||
# 保存失败到文件
|
||
failfile = out_dir / "failed_2dto3d.txt"
|
||
with open(failfile, "w", encoding="utf-8") as f:
|
||
for ident, _, msg in failed:
|
||
f.write(f"{ident}\t{msg}\n")
|
||
print(f"失败分子已保存到: {failfile.resolve()}")
|
||
print("="*60)
|
||
|
||
if __name__ == '__main__':
|
||
main()
|