MolGpka: A Web Server for Small Molecule pKaPrediction Using a Graph-Convolutional Neural Network

Xiaolin Pan, Hao Wang, Cuiyu Li, John Z.H. Zhang, Changge Ji

Research output: Contribution to journalArticlepeer-review


pKais an important property in the lead optimization process since the charge state of a molecule in physiologic pH plays a critical role in its biological activity, solubility, membrane permeability, metabolism, and toxicity. Accurate and fast estimation of small molecule pKais vital during the drug discovery process. We present MolGpKa, a web server for pKaprediction using a graph-convolutional neural network model. The model works by learning pKarelated chemical patterns automatically and building reliable predictors with learned features. ACD/pKadata for 1.6 million compounds from the ChEMBL database was used for model training. We found that the performance of the model is better than machine learning models built with human-engineered fingerprints. Detailed analysis shows that the substitution effect on pKais well learned by the model. MolGpKa is a handy tool for the rapid estimation of pKaduring the ligand design process. The MolGpKa server is freely available to researchers and can be accessed at

Original languageEnglish (US)
Pages (from-to)3159-3165
Number of pages7
JournalJournal of Chemical Information and Modeling
Issue number7
StatePublished - Jul 26 2021

ASJC Scopus subject areas

  • General Chemistry
  • General Chemical Engineering
  • Computer Science Applications
  • Library and Information Sciences


Dive into the research topics of 'MolGpka: A Web Server for Small Molecule pKaPrediction Using a Graph-Convolutional Neural Network'. Together they form a unique fingerprint.

Cite this