search

HaotianZhangAI4Science / ResGen

3D_Molecular_Generation
MIT License
72 stars 11 forks source link

The issue of data #3

Open xueyuanyuan0410 opened 11 months ago

xueyuanyuan0410 commented 11 months ago

Hello, the 4iiy.pdb downloaded from your code is different from the 4iiy.pdb downloaded directly from the pdb database, may I ask what you have done? And how did you convert crossdocked original dataset (pdb file)to this pocket dataset(pdb file with HEADER POCKET)

HaotianZhangAI4Science commented 11 months ago

Sorry, I kind of forgot about the specific process. I just randomly selected an example for illustration. I guess you may need to remove water and other solvent in pymol, and then use pdbfixer to fix the protein's missing parts. The related code can be found at https://github.com/HaotianZhangAI4Science/AI-Physics-DrugDiscovery/tree/main/pdb_blog.

HaotianZhangAI4Science commented 10 months ago

It just occurred to me. The CrossDock dataset is not directly retrieved from PDB database. As its name suggested, this dataset underwent the cross-docking process. More information can be found in its original paper!