Thank you
This live web event has ended. Thank you for attending.
Description
In the process of developing new drugs, finding good or near-optimal solutions to challenges in drug formulation and discovery programs is essential. However, the limitations of time, expenses, and data availability often constrain the options explored in pharmaceutical applications. Here, we explore how certain computer modeling techniques can impact drug development programs by generating valuable data, often in quantities that would be impractical or impossible to obtain otherwise.
Physics and Structure-Based molecular modeling (PSB) plays utilizes the 3D structure of molecules and their interactions to generate data through a variety of techniques, including energy minimizations, conformation searches, docking, transition state prediction, and molecular dynamics simulations. Machine Learning (ML) techniques construct correlations between experimental or calculated values and more complex quantities. The impact of both PSB and ML has been steadily growing for decades within the pharmaceutical industry.
One of the main challenges in the productive application of ML lies in obtaining a sufficiently large and diverse set of training data to address the challenges at hand. However, an strong synergy emerges when PSB calculations are practical and reliable enough to systematically augment or generate large volumes of data for use in creating ML models.
This talk will present selected usage cases for each of these techniques and their synergies, showcasing their application in important areas such as:
- Crystal polymorph prediction
- API solubility
- pKa prediction
- Structure development and evolution in pharmaceutical systems
- Protein viscosity calculations
Learning Objectives:
- Understand the limitation of experimental data encountered in drug product and manufacturing process development with respect to API polymorphs and the resulting risks
- Identify projects for which computation tools can contribute to de-risking during final API form selection
- Combine the results from experiments and computations to gain the most insight into manufacturing issues related to a crystal polymorph