fluoriclogppka

Tool for pKa, logP prediction


Keywords
pKa, logP, tool
License
MIT
Install
pip install fluoriclogppka==0.2.1

Documentation

FluoricLogPpKa

Tool for predicting logP and pKa values.

Features

  • Obtain a lot of features from the molecule using custom methods, rdkit and mordred.
  • Predict pKa and logP values using H2O models.

Installation

FluoricLogPpKa requires installed Java to run.

pip install fluoriclogppka

How to use to predict logP:

import fluoriclogppka

if __name__ == "__main__":

    SMILES = "F[C@H]1C[C@H](F)CN(C1)C(=O)C1=CC=CC=C1"

    inference = fluoriclogppka.Inference(SMILES=SMILES,
                                         target_value=fluoriclogppka.Target.logP)
        
    inference.predict()

How to use to predict pKa:

import fluoriclogppka

if __name__ == "__main__":

    SMILES = "FC1(F)CC(C(O)=O)C1"

    inference = fluoriclogppka.Inference(SMILES=SMILES,
                                         target_value=fluoriclogppka.Target.pKa)
        
    inference.predict()

Experimental tool for prediction (H2O models):

import fluoriclogppka

if __name__ == "__main__":

    SMILES = "FC1(F)CC(C(O)=O)C1"

    inference = fluoriclogppka.Inference(SMILES=SMILES,
                                         target_value=fluoriclogppka.Target.pKa,
                                         model_type=fluoriclogppka.ModelType.h2o)
        
    inference.predict()