I5*
Quantum Mechanical Scoring for Computational Structure-based Drug Design
Dr Adam Pecina, Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences; Dr Jan Řezáč, Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences; Dr Jindřich Fanfrlík, Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences; Dr Martin Lepšík, Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences
IOCB Tech
Challenge
Computer-aided drug design aims to streamline the costly and time-demanding drug discovery and development process. Accurate prediction of protein-ligand binding affinities (scoring) is the cornerstone of all the steps from virtual screening to hit-to-lead optimization. Common scoring functions are quick (seconds per compound) but lack both accuracy and reliability. The state-of-the-art free-energy methods yield better predictions but still rely on the variable accuracy of molecular mechanics force fields and are much slower (days). There is an urgent need for new fast and accurate binding affinity prediction methods.
Technology
Our technology solves the key step in computer-aided drug design, i.e., reliable estimation of the binding affinity of drug candidate (small organic molecule called ligand) to its therapeutic target, usually a protein. We have designed a novel scoring function based on our own quantum-mechanical methods. These offer a favorable accuracy to cost ratio and are applicable to diverse ligand chemistries. Our scoring function achieves accuracy comparable to the state-of-the-art methods but is much faster (minutes). As a purely physics-based approach, it can handle any systems for which a structure is available, and, unlike more empirical approaches and machine learning methods, does not depend on any system-specific data or prior knowledge. We have demonstrated its power in multiple studies covering diverse use cases ranging from virtual screening to detailed ranking of active compounds and their proposed modifications.
Commercial Opportunity
Our scoring function can be readily integrated into computer-aided drug design workflows used in the pharma industry. The viability of our approach had already been demonstrated in practice it had been licensed (non-exclusively) to two of TOP 10 pharmaceutical companies and these licenses are being extended repeatedly. In addition to licensing, we aim to provide this technology together with our unique know-how as a service.
Development Status
The technology is implemented in a fully functional software automating the whole complex computational protocol. Further methodological development and validation is actively pursued by our team at the Institute of Organic Chemistry and Biochemistry in Prague. Simultaneously, we are heading towards founding a spin-off company.
Patent Situation
Our main asset is the software implementing the scoring function, legally protected by the software license and copyright.
Further Reading
1. Pecina A., Fanfrlík J., Lepšík M., Řezáč J., SQM2. 20: Semiempirical quantum-mechanical scoring function yields DFT-quality protein–ligand binding affinity predictions in minutes. Nature Communications 2024, 15, 1127. 2. Pecina A. et al., SQM/COSMO Scoring Function: Reliable Quantum-Mechanical Tool for Sampling and Ranking in Structure-Based Drug Design. ChemPlusChem 2020, 85, 2361. 3. Pecina A. et al., Ranking power of the SQM/COSMO scoring function on carbonic anhydrase II–inhibitor complexes. ChemPhysChem 2018, 19, 873. 4. Pecina A. et al., The SQM/COSMO filter: reliable native pose identification based on the quantum-mechanical description of protein–ligand interactions and implicit COSMO solvation, Chemical Communications 2016, 52, 3312. 5. Lepšík M. et al., The Semiempirical Quantum Mechanical Scoring Function for In Silico Drug Design. ChemPlusChem 2013, 78, 921. 6. Fanfrlík J. et al., A reliable docking/scoring scheme based on the semiempirical quantum mechanical PM6-DH2 method accurately covering dispersion and H-bonding: HIV-1 protease with 22 ligands. The Journal of Physical Chemistry B 2010, 114, 12666.
