A new spin-out university-based company has given drug manufacturers the opportunity to use computational chemistry to create “designer molecules” for the first time in an industrial setting.
The company, Quantum Genetics, is a spin out from the UK’s Northumbria University and offers manufacturers a unique and innovative solution for identifying new catalysts and reactants. Instead of using computational chemistry as a way of interpreting lab-based results during the creation of new molecules , the new process, known as Quantum Directed Genetic Algorithms™ (QDGA™), uses computational chemistry to predict as-yet-unimagined new molecules with exactly the right properties for an industrial application.
“The process is conducted primarily in the computer and intellectual domains,” says Dr Marcus Durrant, pioneer of QDGA and Technical Director of Quantum Genetics. “This means that as well as focusing on the chemistry of a required application, traits can also include almost anything inside and outside normal experimental domains, such as commodity-price and volatility, solubility, toxicity and weight. This ability to consider a wide set of key issues in the early phases of product development can have a radical and disruptive impact on the cost of bringing new products to market.”
The benefits of QDGA over traditional lab-based methodology are as follows.
● more objective
● novel solutions that will be highly disruptive of the status quo
● creates products with strong patent potential
● significant reductions in development costs
● radical reduction in time-to-market
● increased confidence in the feasibility of a new product
● complements existing in-silico methods
Two multinational companies have already commissioned commercially confidential work from Quantum Genetics, and, although these projects lie outside the synthesis area, Dr Durrant foresees the QDGA will have a number of applications that would be of interest to drug manufacturers.
“The great strength of QDGA is that it is can be tailored to very specific problems,” he says. “For example, in the case of a generic, off-the-shelf asymmetrical catalyst, QDGA could be used to refine the structure of the catalyst so it was more closely targeted to the desired transformation and so perhaps deliver cost savings.
“On the other hand, it could also be used to create new catalysts for processes for which there is as yet no commercially valid route. Let’s say a new high-potential drug has been produced in the laboratory or been found in nature, but that there is no catalyst available to scale-up production to industrial levels. If you had some idea of the chemistry you needed, QDGA would allow you to search for the catalyst you required. This of course opens up the possibility of QDGA being instrumental in bringing drugs to market that might otherwise not be economically viable.
“Another possibility for QDGA is to use it in drug development. For example, if you have an enzyme or receptor protein structure and you want to find something that binds to the active site, QDGA would not only let you to look for such ligands, but would also allow you to factor in other things such as solubility, how it would partition between fat and blood, and, possibly, even toxicity. There are some circumstances, such as in the case of poorly characterized but efficacious herbal remedies, in which we could even start from a position of no knowledge of the structure. In this mode, QDGA becomes a kind of data mining tool.”
Although QDGA’s current contracts are confidential, an aspect of the process that is already in the public domain is a prediction of the function of an enzyme produced by the fungus Rhizopus oryzae, This fungus has been used in a number of drug metabolism studies with benefits for osteoporosis therapy and the treatment of various vascular disorders. However, the actual structure of the enzyme has proved difficult to establish with traditional methods and the location of the relevant binding site on the molecule has never been worked out. By using the QDGA process, Dr Durrant’s team has been able to develop a 3D structure-function model for the enzyme which allows accurate predictions to be made about its performance.
The inspiration for QDGA comes from Dr Durrant’s research into the use of quantum calculations to model biological evolution. However, unlike natural evolution, which is a random and open-ended process, QDGA is controlled by the scientist and is directed to create a new molecule with designed-in traits. It is in a sense a process for creating “designer” catalysts and reactants rather than serendipitously discovering naturally occurring or pre-supposed and manufactured products.
Using a proprietary study method and software, QDGA creates something akin to the process of mutation and selection that is seen in the biological world. First, the Quantum Genetics team researches all known information about the qualities the final molecule will need and selects candidate molecules that have the potential to evolve in the desired direction. It may be that the availability of prior research is quite limited. Indeed, the process can be initiated with randomly selected candidates with none of the desired traits. The most suitable molecules that emerge from the virtual environment are then selected and the process run again and again until the optimum solution emerges. QDGA can thus be used either to improve existing processes or to create brand new products.
Quantum Genetics is a major success out of Northumbria University’s High Performance Business Development Programme. This matches commercial and academic skills with development funding to create spin-outs that are built on the research base at the University.
-The author is Senior Press Office, Northumbria University