Possible Rotation Projects:
- Develop and evaluate machine learning methods for protein–ligand scoring, docking, virtual screening, or binding affinity prediction.
- Apply structure-based modeling, docking, molecular dynamics, or free-energy approaches to a therapeutic target of interest to identify or prioritize candidate ligands.
Training Technologies Used:
Molecular docking
Virtual screening
Structure-based drug design
Pharmacophore modeling
Molecular shape search
Molecular dynamics
Binding free-energy methods
Cheminformatics
Machine learning and deep learning
Generative modeling
Python; C++; CUDA/GPU computing
High-performance and high-throughput computing
Web-based scientific software development
Open-source software engineering.
- B.S. Computer Science, Carnegie Mellon University-2001
- M.S. Computer Science, Carnegie Mellon University-2006
- Ph.D. Computer Science, Carnegie Mellon University-2009
- Postdoc Fellow, Computational Biology, University of Pittsburgh-2011
McNutt AT, Francoeur P, Aggarwal R, Masuda T, Meli R, Ragoza M, Sunseri J, Koes DR. GNINA 1.0: molecular docking with deep learning. Journal of cheminformatics. 2021 Jun 9;13(1):43.
Sunseri J, Koes DR. Pharmit: interactive exploration of chemical space. Nucleic acids research. 2016 Apr 19;44(W1):W442-8.
Ragoza M, Masuda T, Koes DR. Generating 3D molecules conditional on receptor binding sites with deep generative models. Chemical science. 2022;13(9):2701-13.
Dunn I, Koes DR. FlowMol3: flow matching for 3D de novo small-molecule generation. Digital Discovery. 2026;5(5):2052-66.
Dunn I, Toft L, Katz T, Gupta J, Shah R, Hettiarachchi R, Koes DR. OMTRA: A Multi-Task Generative Model for Structure-Based Drug Design. 2025 Workshop on Machine Learning in Structural Biology. Dec 2025.
Dunn I, Pirhadi S, Wang Y, Ravindran S, Concepcion C, Koes DR. CACHE Challenge# 1: Docking with GNINA Is All You Need. Journal of Chemical Information and Modeling. 2024 Dec 10;64(24):9388-96.
Gau David, Lewis T, McDermott L, Wipf P, Koes DR and Roy P. Structure-based virtual screening to identify first-generation inhibitor of profilin:actin interaction with anti-angiogenic property. J. Biol. Chem. 2018 Feb 16;293(7):2606-16. doi:10.1074/jbc.M117.809137.
Jia H, Brixius B, Bocianoski C, Ray S, Koes DR, Brixius-Anderko S. Deciphering the Role of Fatty Acid–Metabolizing CYP4F11 in Lung Cancer and Its Potential As a Drug Target. Drug Metabolism and Disposition. 2024 Feb 1;52(2):69-79.
Bhingarkar A, Wang Y, Hoshitsuki K, Eichinger KM, Rathod S, Zhu Y, Lyu H, McNutt AT, Moreland LW, McDermott L, Koes DR, Fernandez CA. Duvelisib is a novel NFAT inhibitor that mitigates adalimumab-induced immunogenicity. Frontiers in Pharmacology. 2025 Jan 9;15:1397995.
My research group develops novel computational algorithms and full-scale software systems to support rapid, inexpensive, and accessible drug discovery. We focus on structure-based drug design, molecular docking, virtual screening, pharmacophore and shape-based search, machine learning for protein–ligand interactions, generative molecular design, and open-source tools that make these methods broadly usable by the biomedical research community.
We aim to connect methods development with real-world therapeutic discovery. We build tools such as GNINA, Pharmit, and related molecular modeling infrastructure, and we apply these methods prospectively to identify and optimize small molecules for challenging biological targets. The lab is especially interested in integrating machine learning, deep neural networks, cheminformatics, molecular modeling, and scalable computing to remove practical barriers to computational drug discovery.
