The Potential of AI in Tailoring Artificial DNA for Drug Development

Artificial DNA tailor-made for future drug development

Researchers at Chalmers University of Technology in Sweden have successfully designed synthetic DNA that controls cell protein production with the aid of AI. This technology could help develop and produce vaccines, drugs to treat severe diseases and alternative food proteins at a much lower cost than currently. The expression of our genes is fundamental to all living organisms. The genetic code of DNA is translated into the messenger RNA molecule (mRNA), telling the cell factory what protein to make and in what quantities.

Researchers have spent a lot on trying to control gene transcription because, among other things it can contribute to the creation of protein-based medicines. Recent examples include the mRNA against Covid-19 which told the cells of the body to produce the protein on the surface the coronavirus. The immune system of the body could then be taught to produce antibodies against virus. It is also possible to train the immune system of the body to fight cancer or other complex diseases by understanding the genetic code that produces specific proteins. The majority of new drugs today are protein-based. However, the methods for making them are expensive and slow because it’s difficult to control the way DNA is expressed. A research group led by Aleksej Zelezniak, Associate professor of Systems Biology at Chalmers, made a major breakthrough last year in controlling the amount of protein produced from a DNA sequence.

\”First, it was about the ability to’read’ all of the instructions in a DNA molecule. \”We have now succeeded in creating our own DNA which contains exact instructions on how to control the amount of a particular protein,\” Aleksej Zelezniak says about the latest breakthrough by the research group. The new method uses a similar principle to that used by AIs when they create faces that resemble real people. The AI learns how to create a wide variety of faces and then can create new, natural-looking ones. The AI can then modify a facial feature by saying, for instance, that it should have a new hairstyle or look older. It would have taken much longer and been more difficult to program a realistic face without AI. The researchers’ AI was also taught the DNA regulatory code and structure. The AI designs synthetic DNA where the regulatory information can be easily modified in order to achieve desired gene expression.