HannesStark / EquiBind

EquiBind: geometric deep learning for fast predictions of the 3D structure in which a small molecule binds to a protein
MIT License
479 stars 109 forks source link

the original purpose and actual effect of reduce_receptors.py script file #72

Open tangmaomao16 opened 2 months ago

tangmaomao16 commented 2 months ago

I read the source code of the script file https://github.com/HannesStark/EquiBind/blob/main/data_preparation/reduce_receptors.py

I think the core coding is f"reduce -Trim {rec_path} > {os.path.join(data_path, name, f'{name}_protein_obabel_reduce_tmp.pdb')}" f"reduce -HIS {os.path.join(data_path, name, f'{name}_protein_obabel_reduce_tmp.pdb')} > {os.path.join(data_path, name, f'{name}_protein_obabel_reduce.pdb')}"

Then I search for the "reduce" command line tool document, and I find https://github.com/rlabduke/reduce/blob/master/README.usingReduce.txt

-Trim remove (rather than add) hydrogens -HIS create NH hydrogens on HIS rings

I want to ask the authors of EquiBind. Do you want to remove most hydrogen atoms in the protein PDB file to decrease the computation of the program? But why still create NH hydrogens on HIS rings?

Please tell me the original purpose and actual effect of reduce_receptors.py script file. Thank you.