US Pharm. 2024;49(10):3.

Artificial intelligence (AI) has the potential to accelerate the process of identifying new molecules for mental health disorders drug development, according to a recent study by scientists from Uppsala University in Sweden. By predicting the three-dimensional structures of key receptors, the authors of the research in Science Advances contend, AI could speed up the drug-development process.

In medication development, experimental methods are often employed to determine the structures of target proteins and to understand how molecules bind to them. This information is required to efficiently design drug molecules, but the experimental methods to determine these structures can be demanding and time-consuming.

Advancements in AI technology, however, now allow for more accurate predictions of protein structures. In the study, the researchers employed AI to model the structure of the TAAR1 receptor, a target protein for developing drugs to treat mental health conditions such as schizophrenia and depression. By utilizing supercomputers to search through chemical libraries, they identified molecules that could potentially bind to the receptor.

Experimental testing confirmed that many of these molecules activated the receptor, with promising results in animal tests. The researchers were surprised by the accuracy of the AI-generated structures, which outperformed traditional methods. Molecules predicted to bind to the receptor were then tested in experiments by research colleagues at Karolinska Institutet. An unexpectedly large number of the molecules activated TAAR1, and one of the most potent also showed promising effects in animal experiments. During the final stage of the study, experimental structures for TAAR1 suddenly became available, and the researchers were able to compare them with the AI models.

“The accuracy of the structures generated with AI was astonishing—I couldn’t believe it. The results also show that modeling with AI is significantly better than traditional methods. We can now use the same strategy for receptors that we previously could only dream of working with,” explained Jens Carlsson, who led Uppsala University’s role in the study.

Owing to the development of AI methods, the scientists said, the structures of proteins can now be predicted with higher accuracy—potentially allowing researchers to target previously inaccessible protein molecules.

This issue of U.S. Pharmacist details a number of new drugs, including resmetirom (the first drug of its kind engineered to treat advanced metabolic liver disease); etrasimod for moderately to severely active ulcerative colitis; berdazimer for molluscum contagiosum; cefepime and enmetazobactam for complicated urinary tract infections; letibotulinumtoxinA-wlbg for moderate-to-severe frown (glabellar) lines; and bimekizumab-bkzx for moderate-to-severe plaque psoriasis.

The content contained in this article is for informational purposes only. The content is not intended to be a substitute for professional advice. Reliance on any information provided in this article is solely at your own risk.

To comment on this article, contact rdavidson@uspharmacist.com.