April 5, 2018
New liquid modeling technique predicts chemical reactions and lowers drug development costs
WEST LAFAYETTE, Ind. – A Purdue-affiliated company is developing a way to reduce drug development costs by more accurately and efficiently modeling molecules and chemical reactions in liquid solutions. This will allow chemists to better understand process details of molecule synthesis.
QUAIL Modeling LLC, an acronym for Quantum Applications in Liquids, was co-founded by Tillmann Kubis, research assistant professor in Purdue's School of Electrical and Computer Engineering, Network for Computational Nanotechnology and Purdue Center for Predictive Materials and Devices, and James Charles, a Ph.D. student in the same department.
The software was developed out of a need to better understand how molecules react in liquids. A video about the technology can be viewed here.
“When you have a molecule that’s expected to behave a certain way, current models allow you to predict its behavior in the vacuum only where the molecule is basically isolated. However, drugs are supposed to interact with a liquid surrounding such as blood. So far, there is no way to reliably predict molecular behavior in liquids, where the drug will actually take effect,” Kubis said. “The first question we will answer is how will these molecules change when they are put into a liquid surrounding such as a patient’s bloodstream.”
Kubis said modeling liquids is a big and still not fully solved challenge in quantum chemistry.
“It is not yet understood how to model water and how to model molecules when they are dissolved,” he said. “Molecules in an aquatic surrounding face too many chaotic perturbations for the state-of-the-art quantum models. Typical quantum descriptions are not able to efficiently handle such intense uncertainties.”
QUAIL Modeling is expanding the Non-Equilibrium Green Function method (NEGF) to the realm of liquid quantum chemistry. This method will allow chemists to calculate time-dependent non-equilibrium expectation values such as current and densities, energy exchange and entropy changes of the system. The NEGF method is already a widely accepted method in the electrical engineering and high-energy physics world.
“One of the holy grails of quantum chemistry is the prediction of the solvation energy, i.e., the energy change when a molecule dissolves in a liquid.” Kubis said. “QUAIL is working directly to solve this problem. In spite of its importance, this problem has been unsolvable so far. We tackle it by combining the quantum effects with the statistical uncertainties of a liquid environment. We can do this explicitly for any kind of liquid and dissolved molecule.”
Kubis said this method will significantly reduce the cost of drug development.
“The potential of this is gigantic. There are only about 20 drugs released to the market each year and it costs approximately $5-12 billion to get each of them to that stage,” he said. “Decreasing these expenses by even only 10 percent can make a huge difference.”
Kubis said it is vital for companies to test molecules in perfect purity, without any synthesis-by-products or free from undesired molecular chirality. This is essential to determine any adverse side effects of the actual drug molecule. No impurities produced during the drug’s synthesis may blur that information.
“When big companies have a molecule to synthesize, they have their own service providers with databases that typically give them about 15-20 different reaction paths that could yield a high concentration, or high purity, of the desired molecule. The high costs come then in thoroughly testing each reaction path. Our technology could narrow down the 20 reactions to much fewer, more accurate candidates, or even identify new reaction paths that are not on the radar of the incomplete databases,” he said. “This will reduce drug development costs and increase the reliability of drug tests.”
Technology used by QUAIL Modeling is licensed through the Purdue Research Foundation Office of Technology Commercialization. The company is a member of the Purdue Startup Class of 2017.
The company is working on academic theories with a focus toward industrial applications.
“Our first goal is to correctly predict the solvation energy,” Kubis said. “Modeling chemical reactions is our long-term goal; we need a lot of development to complete that.”
The development of this technology has been previously supported academically by the Center of Materials and Predictive Devices at Purdue. Kubis said QUAIL Modeling is currently seeking partnerships and funding.
“We are seeking out partnerships within industry, as well as funding and guidance in what are the most urgent open questions to pharmaceutical and chemical industry,” he said. “We need to discuss the open questions with the specialists of the field since we are not chemists. Having some guidance will help us stay on target.
James Charles, firstname.lastname@example.org