AI tailors artificial DNA for future drug development
AI tailors artificial DNA. Credit: Yen Strandqvist

With the help of an artificial intelligence, researchers at the University of Technology, Sweden, have succeeded in designing synthetic DNA. The technology can contribute to the development and production of vaccines, drugs for severe diseases, and alternative food, at much lower costs than today.

The function of cells is dependent on the expression of genes. The genetic code in DNA is transcribed to a molecule called mRNA, which is used to tell the cell's factory which to make and which to discard.

Researchers have put a lot of effort into trying to control the expression of genes because it can contribute to the development of drugs that are related to the human body. The body's cells were instructed to make the same molecule found on the surface of the coronaviruses in the vaccine.

The body could learn to fight the virus. It is possible to teach the body's immune system to destroy cancer cells and other complex diseases by understanding the genetic code behind their production.

It is difficult to control how the DNA is expressed in order to make a new drug. An important step was taken last year by a research group led by the Associate Professor of Systems Biology.

The first thing it was about was being able to read the instructions. According to the research group's latest important breakthrough, we have succeeded in designing our own DNA that contains the exact instructions to control the quantity of a specificProtein

The genes were made to order.

The principle behind the new method is the same as the one used in artificial intelligence. The artificial intelligence can create new faces by learning what a large selection of faces look like. It is easy to make a face look older or have a different hairstyle.

Without the use of artificial intelligence, programming a believable face from scratch would have been much more time consuming. The researchers' artificial intelligence has been trained on the structure and regulatory code of DNA. The artificial intelligence is able to modify its regulatory information in the desired direction of genes. The artificial intelligence is told how much of a gene is desired and then "prints" the appropriate sequence of genes.

It's a long and complex molecule. It's very difficult to make changes to it by iteratively reading and changing it. It can take years of research to find a solution. It's better to let an artificial intelligence learn how to navigate. Jan Zrimec, a research associate at the National Institute of Biology in Slovenia, is the author of a new book.

The yeast Saccharomyces cerevisiae has cells that look like mammals. Human cells are the next step. Researchers hope that their progress will have an impact on the development of new and existing drugs.

It can take many years for a drug to be developed for a disease. It's not possible to get a return on investment for some of them. "With our technology, it is possible to develop and manufacture a lot more efficiently so that they can be marketed."

The research was published in a peer-reviewed journal.

Jan Zrimec and her team control gene expression with deep generative design of regulatory DNA. There is a DOI titled " 10.1038/s41467- 022-32818-8".

Journal information: Nature Communications