search

gautammalik-git / AutoDock-GPU-Pipeline

This pipeline facilitates setting up ligand docking against a protein using AutoDock-GPU. It streamlines the process of docking a ligand library onto a protein structure, leveraging the enhanced performance of AutoDock-GPU for faster results.
MIT License
3 stars 0 forks source link

energy_file = "checklines.txt" #1

Open wen1234defeng opened 4 months ago

wen1234defeng commented 4 months ago

What file is this "checklines.txt"? How did you get it?

gautammalik-git commented 4 months ago

Hey! I appreciate you investigating the matter. Essentially, it's a text document containing all compound names along with their binding affinities. It ought to be named 'Energy.txt' since the 'binding_energy.py' script saves the energies in that file. Thanks for catching the error. I've made the necessary updates to the 'full_dir.py' file.

wen1234defeng commented 4 months ago

嘿！感谢您调查此事。本质上，它是一个包含所有化合物名称及其绑定亲和力的文本文档。它应该命名为“Energy.txt”，因为“binding_energy.py”脚本将能量保存在该文件中。感谢您发现错误。我已经对“full_dir.py”文件进行了必要的更新。

After I changed it to Energy.txt, the extracted "full_dir" directory was empty.

gautammalik-git commented 4 months ago

The purpose of the _fulldir.py script is to duplicate the dlg files of the ligands into a new directory. This allows you to extract coordinates from these files to obtain a protein-ligand complex. In the _fulldir.py file, there's a variable named 'dir', initially set as 'dir = lines[6:13]'. It should be adjusted to 'dir = lines[:13]'. Alternatively, you can directly download the updated file from the repository.

wen1234defeng commented 4 months ago

The purpose of the _fulldir.py script is to duplicate the dlg files of the ligands into a new directory. This allows you to extract coordinates from these files to obtain a protein-ligand complex. In the _fulldir.py file, there's a variable named 'dir', initially set as 'dir = lines[6:13]'. It should be adjusted to 'dir = lines[:13]'. Alternatively, you can directly download the updated file from the repository.

OK! Thank you very much. I have a question for you. When I was sorting out the ad_gpu docking results, I found that the small molecule was protonated, resulting in an abnormally low affinity (-30kcal/mol). This result bothers me very much, and I don’t know whether to keep it. SMILES C=C(CNC(=O)[C@@H]1CCCC[NH2+]1)OC@Hcc1)[C@@H] (CF)NC(=O)C(Cl)Cl

gautammalik-git commented 4 months ago

You can check the protonation state using different software, just set the pH to 7.4. If the protonation makes sense biologically, then keep the molecule. Also, try docking with the uncharged version to see if the charge is affecting the result.