About pHdockUI
Our Mission
We're advancing computational drug discovery by accounting for pH-dependent molecular behavior. Traditional docking tools often overlook protonation states, leading to inaccurate predictions. Our suite bridges this gap with state-of-the-art machine learning and quantum-informed models.
Our Inspiration
The project emerged from observing critical failures in drug discovery pipelines where promising candidates failed due to incorrect protonation state modeling. At physiological pH, many drug molecules exist in multiple protonation states, each with different binding affinities.
Inspired by recent advances in graph neural networks and the availability of large-scale pKa datasets, we developed an integrated approach that combines:
- Fast, accurate pKa prediction using ensemble ML models
- Quantum mechanical validation for challenging cases
- Seamless integration with popular docking tools
- User-friendly interfaces for both researchers and educators
Meet Our Team
Passionate researchers dedicated to advancing computational chemistry
Ravindra Lakkireddy
Co-founder, CTO
Head of training, scoring, and data collection, handling abalation studies as well as quantum integration
Gianluca Radice
Entry-level ML Intern
Defined the frameworks for the GCNNs used, lead efforts for constructing a full pipeline from SMILE to website output
Denis Motuzenko
Co-founder, COO
Conceptual and chemical leader, engaged with the interpretation layer and organized quantum logic
Academic Affiliations
Science, Math, and Computer Science Magnet Program (SMCS)
Acknowledgments
We thank our advisors, collaborators, and the open-source community for their invaluable contributions. Special thanks to the developers of RDKit, PyTorch Geometric, and AutoDock for providing the foundation upon which this work builds.