mayrf/pkasolver

toolkit for prediction pKa values of small molecules via graph convolutional networks

pkasolver's Issues

Cannot reproduce the result in `pka_prediction.ipynb` notebook

Hello, developer @mayrf

I just run pka_prediction.ipynb on Google Colab. But I got different result with your original version.

Can you tell me more document for how to use this package?

Thanks for any help.

pka solver stops running on large librairies of compounds

Hi,

I've been trying to run pka solver (calculate microstates function) on a fairly large dataset of compounds (>1000). Unfortunately, it seems the function simply stops running after a while. Other similarly large datasets give the same problem, and stop after a seemingly random amount of compounds, with no error message.

Happy to share the code I'm using if that is helpful.

Cheers

Travis CI Security Breach Notice

MolSSI is reaching out to every repository created from the MolSSI Cookiecutter-CMS with a .travis.yml file present to alert them to a potential security breach in using the Travis-CI service.

Between September 3 and September 10 2021, the Secure Environment Variables Travis-CI uses were leaked for ALL projects and injected into the publicly available runtime logs. See more details here. All Travis-CI users should cycle any secure variables/files, and associated objects as soon as possible. We are reaching out to our users in the name of good stewards of the third-party products we recommended and might still be in use and provide a duty-to-warn to our end-users given the potential severity of the breach.

We at MolSSI recommend moving away from Travis-CI to another CI provider as soon as possible. The nature of this breach and the way the response was mis-handled by Travis-CI, MolSSI cannot recommend the Travis-CI platform for any reason at this time. We suggest either GitHub Actions (as is used from v1.5 of the Cookiecutter-CMS) or some other service offered on GitHub.

If you have already addressed this security concern or it does not apply to you, feel free to close this issue.

This issue was created programmatically to reach as many potential end-users as possible. We do apologize if this was sent in error.

KeyError: 'GIN_p.convs.0.nn.lins.0.weight'

KeyError: 'GIN_p.convs.0.nn.lins.0.weight'
i am using the the notebook provided, it errors out protonation_states = calculate_microstate_pka_values(mol, only_dimorphite=True)

Not sure there seems to be a mismatch with the model provided and the code

Using subprocess to run dimorphite_dl

Hello,

I've noticed that Dimorphite-DL is called via subprocess, although it is a python script and can probably be called directly as an imported package, which should be more efficient.

I can try to change it myself if this makes sense.

Discrepancy in protonation site indices with Dimorphite-DL

Hello,

First of all, thank you for making this tool open-source.

I tried to run pkasolver-light on Novartis dataset with Dimorphite-DL. However, it seems that the protonation sites identified by Dimorphite are very different from Marvin's.

When I evaluate the model on the Novartis dataset you provided (with protonation sites identified by Marvin), the RMSE score is around 1.2 on pkasolver-light but with Dimorphite-DL, it is 3.9.

Is it a mistake on my side or are the scores with Dimorphite-DL indeed so low?

Colab Notebook Doesnt Work

For some reason Colab notebook doesnt work.

Reference bug

Edge attributes seem to be always the same for prot/deprot forms of molecules

I have started to add tests for some of the graph generating functions and come across some surprising behavior in one of the added test. It seem as if the edge attributes are the same for prot/deprot forms of the molecule. Can you please double check this @mayrf ? Is this expected behaviour?

A test case that checks for this is here:
https://github.com/MayrF/pkasolver/blob/bf01b4be71b7aa97a5984bdae75502e81f3b228f/pkasolver/tests/test_data_generation.py#L178

Update test_env.yaml

Hi Fritz!

Could you please update the pkasolver/devtools/conda-envs/test_env.yaml and include the packages that you have installed to run you tools?

thank you and best regards,