Molecular Crystal Polymorph Prediction: High Accuracy at Lower Computational Cost

University Of California-Riverside, Riverside CA

Investigators

Abstract

Professor Gregory Beran of the University of California-Riverside is supported by an award from the Chemical Theory, Models and Computational Methods program in the Chemistry Division to develop new computational tools that will facilitate the prediction of three-dimensional crystal structures. Knowledge of molecular crystal structures is essential in pharmaceuticals and many other areas of chemistry. Different crystal packing arrangements, or “polymorphs,” of the same molecule can exhibit vastly different properties. The occurrences of undesirable polymorphs have caused major drug recalls and other serious problems for patients and pharmaceutical manufacturers in the past. The pharmaceutical industry increasingly employs crystal structure prediction to complement their experimental drug formulation efforts and to reduce the potential for polymorphism-related problems. It has recently been discovered that the current theoretical models in widespread use exhibit significant problems for predicting the crystal structures of drug molecules. This project is developing new computational models that correct these weaknesses and improve the reliability with which crystal structures can be predicted. Software developed by this project will be released to the community as free, open-source software. Beyond the core research, Professor Beran is actively involved in pedagogical efforts to help train next-generation scientists, a large proportion of whom come from low-income, first-generation, and/or traditionally underrepresented minority demographics. This research occurs in three parts. First, new computationally-practical electronic structure methods for modeling the non-covalent interactions that govern molecular conformation and crystal packing are being developed to enable identification of good initial crystal structures. Second, an approach that combines the strengths of the new electronic structure methods for describing intramolecular interactions with the lower computational costs of density functional theory for modeling intermolecular interactions is being developed to enable improved crystal structure predictions in pharmaceutical compounds. Finally, new lower-cost approximations for handling the vibrational contributions to crystalline stability are being investigated to allow investigation of how temperature affects polymorph stability. This project is developing new computational models that correct weaknesses in current theoretical models and improve the reliability with which crystal structures can be predicted. Software developed by this project will be released to the community as free, open-source software. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.