Molecular Modeling and Data Science for De-risking and Cost- saving of Pharmaceutical Development
Introduction
Advances in computational materials science and process simulation have made computational tools essential1,2 in drug development and manufacturing. Integrating computational and experimental approaches helps de-risk pharmaceutical development, leading to significant time and cost savings. This publication explores the scope of molecular modeling and data science (DS) applications at Porton J-STAR (https://www.portonpharma.com/en), highlighting their critical role in guiding drug development, optimizing decision-making, and minimizing experimental risks.
2. Computational Capabilities and Approaches
Porton J-STAR utilizes a wide range of computational methods - from molecular mechanics to quantum chemistry and DS - to study systems ranging from small (MW 300-800 Da) molecules to new modalities (MW 100-4500 Da) in gas, liquid, and solid phases.
Molecular simulation and DS approaches for combined computational and experimental project support are divided into four categories: virtual screening approaches, properties characterization and optimization, chemical reaction and reactivity predictions, and AI/ML-based DoE calculations.
2.1. Virtual screening approaches
Virtual screening is a computational technique used to rank a database of compounds (solvents, coformers, counterions or excipients) to identify the most promising solutions for a targeted experimental follow up.3-5 Typical virtual screening tasks include solubility and solid solvate propensity prediction; cocrystal/salt screening; solvent and solid form screening for impurity rejection; excipient selection for crystallization suppression in amorphous phase.
2.2 Properties characterization and optimization approaches
The primary goal is to predict the characteristic properties of solid forms or molecules either before or instead of experimental measurements, especially when experimentation is infeasible. The following calculations are typically performed: crystal form characterization along salt-cocrystal spectrum;6 mechanical properties prediction; molecular and crystal form analytical properties prediction; crystal shape optimization.
2.3 AI/ML-based DoE
A typical process design task in the pharmaceutical industry is performed in a multidimensional parameter space and requires the optimization of one or more target properties. To reduce the number of experiments needed to optimize multidimensional process, AI/ML-DoE platform is applied.7 It employs a Bayesian optimization approach, beginning with a limited number of initial measurements to guide subsequent experimentation. AI/ML-DoE facilitates rapid and reliable convergence to the parameter space for achieving optimal target properties.
2.4 Chemical reaction and reactivity predictions
Quantum mechanical calculations are conducted to predict heat and free energy of reactions, regio- and stereoselectivity and transition state searches in solvent systems of interest. These calculations can help identify the most effective synthetic pathways before experimentation or predict reaction properties needed for chemical engineering computations.8
3. Conclusions
This publication provided an overview of the advanced molecular modeling and DS technologies that enable the pharmaceutical development process. The combined computational and experimental approaches are instrumental in accelerating the pharmaceutical development process, enhancing accuracy, and reducing both risk and cost.
Connect with Porton J-STAR Experts:
Yuriy Abramov, Executive Director of Computational Chemistry & Data Science:
yuriy.abramov@jstar-research.com
Barbara Thompson, Marketing Manager:
bthompson@portonusa.com
References
- Abramov, Y.A., Ed. Computational Pharmaceutical Solid State Chemistry, John Wiley & Sons, 2016.
- Abramov, Y. A.; Sun, G.; Zeng, Q. J. Chem. Inf. Model. 2022, 62, 1160.
- Abramov, Y.A.; Shah, H. S.; Michelle, C.; Wan, Z.; Xie, T.; Kuang, S. ; Wang, J. Cryst Growth Des 2025, 25, 5210.
- Shah, H.S.; Michelle, C.; Xie, T.; Chaturvedi, K.; Kuang. S.; Abramov, Y.A. Pharm Res. 2023, 40, 2779.
- Abramov, Y.A.; Zelellow, A.; Chen, C.-Y.; Wang,J.; Sekharan, S. Cryst. Growth Des. 2022, 22, 6844.
- Abramov, Y.A.; Wang, J. Cryst. Growth Des. 2024, 24, 4017.
- https://sunthetics.io/
- Lam, Y.; Abramov, Y.A.; Ananthula, R.S., et al. Org. Process Res. Dev. 2020, 24, 1496.
Download '.pdf' Format of the whitepaper.
