Performance Metrics Docking Tasks

[Jonas-Verhellen]

Dear Maintainers,

I have a question regarding the results on the docking tasks in the Tartarus paper. Could you help me understand how the docking scores shown in Tartarus/mols_jupyter/docking.ipynb relate to the performance metrics for protein-ligand design benchmarks shown in Table 3 of the paper?

Kind regards, Jonas

[akshat998]

Hi Jonas,

Thank you for bringing the GitHub issue to our attention.

We previously utilized the mols_jupyter directory to manage our molecules tracking, but we've revised our docking objectives to enhance pipeline stability through improved filtering. Consequently, the data in the Jupyter notebook is now outdated. We will remove this directory to avoid any further confusions (so, this directory is currently unrelated to Table 3).

Thank you for bringing this up, we completely forgot about this. Please let us know if there are any other questions :) Regards Akshat

[Jonas-Verhellen]

Hi Akshat,

Thank you for getting back to me. Could you explain how to run the revised docking objectives?

Running

from tartarus import docking
score_1syh = docking.get_1syh_score(smi)
score_6y2f = docking.get_6y2f_score(smi)
score_4lde = docking.get_4lde_score(smi)

results in the values found in the notebook.

Kind regards,

Jonas

[akshat998]

Hi @Jonas-Verhellen,

The revised docking objectives have already been incorporated into the code. Thus, the code you pasted executes the calculations we proposed in the paper.

It's just that the notebook is outdated and needs to be removed. Besides the notebooks, everything else is completely up to date.

Regards, Akshat

[Jonas-Verhellen]

Hi @akshat998

Okay, thank you for clarifying. In that case, is the output data from the models (the final population of smiles) in the benchmark available somewhere?

Kind regards, Jonas

[akshat998]

Hi @Jonas-Verhellen,

Not at the moment. This can be provided in the future, but not at the moment due to low bandwidth of the team.

Regards Akshat

[Jonas-Verhellen]

Hi @akshat998,

So to summarize, do I understand correctly that the data in the notebooks was incorrect and will be removed, and that the data supporting the results in the paper is unavailable?

Kind regards, Jonas

[akshat998]

Hi Jonas,

To clarify, the data in the notebooks was indeed incorrect and will be removed. The data supporting the results in the paper is temporarily unavailable but will be updated and made accessible soon :)

Best regards, Akshat

[Jonas-Verhellen]

Hi Akshat,

Great! Thanks for all the responses and the final clarification. I'm looking forward to the updated version of the repo, so that I can include the benchmark in an upcoming paper.

Best regards, Jonas

[akshat998]

Hi Jonas,

You're welcome! Please feel free to reach out if you need any assistance with running the benchmark independently of the repo updates. We're here to help if anything is unclear.

Regards, Akshat