The central ideas underlying a specialized simulation technology called free energy perturbation (FEP) date as far back as the 1950’s with the pioneering academic work led by Zwanzig. However, it wasn’t until the early 1980’s when computational chemistry researchers started applying FEP to the calculation of the binding affinity of ligands to their associated receptors. Researchers saw potential for this cutting-edge technology to deepen their understanding of fundamental molecular biology and inform the development of novel drug therapies. However, substantial scientific and technical innovations were required to achieve the level of accuracy, reliability, and throughput needed to leverage this new simulation technology in industrial drug discovery research.
By 2010, decades after those very important initial efforts, the conditions were finally right for developing an accurate and robust FEP methodology that would be useful across many drug discovery projects. One key change was the increasing availability of computing power, including graphical processing units (GPU), to accelerate molecular dynamics simulations, which are the engine of FEP calculations. Trailblazing academic labs also made key basic science advances related to force fields and enhanced sampling algorithms during this time period. Schrödinger thus embarked on a large-scale engineering project to develop a new force field, optimize enhanced sampling techniques that ensure a simulation is able to sample all important drug candidate/protein receptor structures without requiring excessive computational time, develop a user-friendly graphical user interface, and optimize code to efficiently run on GPUs and on the cloud.
While FEP+ is now considered a “gold standard” computational tool for designing high-quality drug candidates, its utility extends far beyond commercial drug design. FEP+ enables uncharted exploration of basic chemistry and molecular biology. For years, Schrödinger’s FEP+ has been largely out of reach of academic researchers, due to the scale of calculations necessary to test out a variety of different hypotheses relevant to their work and the associated costs. Now, with input from the academic community, Schrödinger has developed an affordable offering designed to empower academic researchers to apply FEP+ to basic research activities, including:
- Probing the molecular mechanisms underpinning molecular biology
- Discovering tool compounds with desired selectivity profiles to interrogate novel biology
- Exploring the mechanisms of molecular recognition or signal transduction pathways via protein-protein binding
- Understanding the effects of protein mutations on evolution, diseases, and drug resistance
- Testing and validating methods and hypotheses coming from chemical intuition, molecular simulations, or AI
Academic researchers will now have the ability to deploy state-of-the-art highly predictive FEP calculations at scale. This offering includes all necessary on-demand cloud GPU resources, removing barriers to entry for academic users, since they won’t need to configure GPU clusters in the cloud, set up independent relationships with cloud providers, or become experts in FEP simulation methods themselves. This ease of use and ease of deployment broadens the application of these methods across the academic community, empowering experimentalists to probe fundamental questions of molecular biology — beyond the development of simulation methodologies.
A large portion of our understanding of basic biology continues to emerge through the hard work of an army of academics exploring specialized areas of their field. These basic biology insights and advances form the foundation of many novel drug therapies, as well as our fundamental understanding of life. This shared quest for better medications and deeper knowledge of human biology motivated us to broaden access to FEP+, and we can’t wait to see what new discoveries will result from FEP+ in the academic space.
Note: FEP+ for Academic Research is intended to promote basic research in chemistry, molecular biology, and related fields, and excludes use for commercial purposes such as drug discovery or other IP generating activities.
