Artificial intelligence predicted where new drugs could get close to RNA and DNA

(ORDO NEWS) — Skoltech’s iMolecule research team has developed a solution that predicts the regions of these molecules that are suitable for interacting with suspected drugs based on data on the structure of RNA and DNA. Knowing these so-called binding sites, it is possible to more efficiently and purposefully find formulas for new drugs, including antiviral ones.

The solution presented in the journal Nucleic Acid Research: Genomics and Bioinformatics uses artificial intelligence and determines binding sites more accurately than analogs, since it takes into account the influence of the spatial configuration of the molecule on the accessibility of sites.

For a long time, pharmacologists saw in RNA only an intermediary between the instructions in our genome (DNA) and the functional proteins encoded in them – the proteins themselves remained the target of most drugs. At the same time, it is known that proteins are encoded only in an insignificant part of the 85 percent of the genome from which RNA is transcribed.

The remaining non-coding RNA participates in the regulation of genes or performs other functions, often assuming a certain conformation, that is, a spatial configuration. Since processes involving noncoding RNA can also contribute to the development of diseases, RNA sequences – and DNA too – are increasingly viewed as potential drug targets.

“Nucleic acids, DNA and RNA, are involved, for example, in signaling and other processes that can be medically affected.This approach, in particular, may be suitable for diseases in which disordered proteins or proteins without available binding sites are involved, – explained the head of the study, senior lecturer at Skoltech Pyotr Popov. “In addition to that, there are foreign RNA and DNA, for example viral, – coronavirus, HIV, and so on, which are one of the main targets in the fight against pathogens.”

To unlock the potential of all these putative drug targets, pharmacologists need tools to search through huge databases of chemical compounds – thus establishing which ones interact with a particular nucleic acid and through which sites.

Artificial intelligence predicted where new drugs could get close to RNA and DNA

“Our solution is based on a similar work with proteins,” explained Popov. – Three-dimensional structures of nucleic acids are encoded as high-dimensional tensors. The computer vision algorithm then “looks” at the tensors and looks for regions that look like binding sites.

Once the conformation and the binding site have been detected, targeted drug discovery can begin. Thus, our work is part of the transition from blind brute force to rational drug design . The superiority of the latter becomes more and more noticeable with the growth of connection libraries. ”

One important advantage of the new solution is associated with the fact that DNA and RNA molecules, due to their shape, tend to fold and take different conformations – while changing their properties, including the available binding sites.

Traditional approaches rely on the nucleic acid sequence, that is, the “letter code”, but ignore the conformation, which is a big drawback.

“In addition, most of the previous methods were applicable only to RNA, and specifically to a single strand. And ours works with DNA and two or more chains, and we can even detect sites that appear ‘at the junction’ when several macromolecules interact, ”added Skoltech graduate student Igor Kozlovsky, the first author of the work.

“A good example of why you shouldn’t ignore conformation is associated with the most common type of HIV,” the scientist continued. “It has a region of RNA that many drugs target.

But although the sequence of nucleic acids is the same, changing the conformation of a molecule changes the set of drugs that can act on it. Our neural network predictions reproduce this effect. ”

The new solution has one unexpected application “backwards”: instead of recognizing binding sites on a potential target, you can consider the problematic active substance. It could be a small molecule like a hormone that works to cause illness.

“You can ‘distract’ these small molecules. This requires constructing a short sequence of nucleic acids called an aptamer, which will serve as a target for a problem hormone or other substance.

Obviously, there should be a binding site on the aptamer, and our solution can be used to design aptamers with stronger interactions, Popov concluded.


Contact us: [email protected]

Our Standards, Terms of Use: Standard Terms And Conditions.