A team of researchers from LMU, ETH Zurich, and Roche Pharma Research and Early Development (pRED) Basel has used synthetic intelligence (AI) to expand a cutting-edge approach that predicts the optimal approach to synthesizing drug molecules. “This particularly reduces the number of laboratory experiments needed, thereby extending the power and durability of chemical synthesis,” says David Nippa, leader of the corresponding paper, published in the journal Nature Chemistry. Nippa is a PhD student in Dr. David Konrad’s research organization. at the Faculty of Chemistry and Pharmacy at LMU and Roche.
Active pharmaceutical ingredients usually consist of a design to which functional equipment is attached. These devices allow for an express biological function. To achieve new or improved medical effects, functional equipment is changed and added to new positions in the frame. However, this procedure is especially complicated in chemistry, because the designs, which consist basically of carbon and hydrogen atoms, are themselves unreactive. One way to activate the design is the so-called borilation reaction. In this procedure, a chemical compound containing the detailed boron is attached to a carbon atom in the structure. This boron compound can then be replaced using a variety of medically effective equipment. Borylation has wonderful potential, but it’s tricky in the lab.
Collaborating with PhD student Kenneth Atz at ETH Zurich, David Nippa developed an AI design trained from knowledge of trusted clinical paints and experiments from an automated laboratory at Roche. It can effectively predict the borylation position of any molecule and provides optimal situations for chemical transformation. “Interestingly, the predictions advanced when the three-dimensional data of the raw tissues, and not just their two-dimensional chemical formulas, were taken into account,” he says. Atz.
The approach has already been used effectively to identify positions in existing active ingredients where more active equipment can be introduced. This is helping researchers more rapidly expand new and more effective variants of active ingredients in known drugs.
Story Source:
Material provided through Ludwig-Maximilians-Universität München. Note: Content is subject to replacement in terms of taste and length.
Journal Reference:
Stay informed with ScienceDaily’s free email newsletter, updated daily and weekly. Or check out our many news feeds in your RSS reader:
Stay up-to-date with the latest news from ScienceDaily’s social media:
Let us know what you think of ScienceDaily – we appreciate both positive and negative feedback. Do you have any challenges on this site?