Artificial intelligence will make the stages of new drug discovery faster, better and more productive

Artificial intelligence has the potential to transform how humans learn, work and interact with each other in every aspect of life, including health.
Although the use of AI in drug discovery is still in its infancy, it has the potential to revolutionize the way new drugs are developed.

AI can help speed up the drug discovery process, reduce the cost of drug development, and improve the safety and efficacy of new drugs.

In 2019, scientists at the Massachusetts Institute of Technology in the United States did something unusual in modern medicine when they discovered a new antibiotic called holicin.

In May this year, another group discovered a drug compound called apoucin, which shows beneficial activity as an antibiotic. What distinguishes these two compounds is not only that they can be used against two of the most dangerous antibiotic-resistant bacteria known, but also how they were identified.

In both cases, the researchers used an artificial intelligence model to sift through millions of candidate compounds to determine which would work best against each “superbug,” according to a report published in the journal. “The Economist”.

“Needle in a Haystack”

He pointed out to the paper that the artificial intelligence model looked at the chemical structure of a few thousand known antibiotics and how successful they were (or weren’t) at fighting bacteria in the lab.

During this exercise, the model was able to detect interactions between chemical structures and success in killing bacteria.

Once the AI ​​released its shortlist, scientists tested it in the lab and identified their antibiotics.

If discovering new drugs is like looking for a “needle in a haystack,” says Regina Barzilai, a computer scientist at MIT who helped discover apocycin and hallucin, AI acts like a “metal detector.”

It takes years of clinical trials for drug candidates to move from the lab to the clinic for use against diseases, but there’s no doubt that artificial intelligence has sped up the initial trial-and-error part of the process.

A long and complicated task

Traditionally, discovering and developing new drugs to treat diseases is a long and complex task, but artificial intelligence is helping to speed up this process.

To develop drugs, researchers need to understand the biological and genetic differences that cause diseases.

By applying AI to anonymized medical data sets, such as electronic health records or lab results, scientists can fill in missing information about the causes of these diseases.

AI is helping researchers to develop more targeted drugs, and progressing towards precision medicine.

For example, in oncology, AI algorithms can be applied to digital images of biopsies to help detect subtle differences between tumors that indicate the presence of genetic mutations in a subset of patients.

Researchers can use these findings to tailor drugs to this subgroup of patients.

The same algorithms that help identify genetic mutations can be used to identify these patients in the real world to facilitate clinical trial recruitment and clinical decision making.

“Drug discovery is a very challenging process, with only a small percentage of drug compounds moving into clinical trials and an even smaller percentage becoming approved drugs,” says Chris Moy, Scientific Director of Oncology, Data Science, Digital Health and R&D at Johnson & Johnson. Johnson.

He adds: “AI will not only help us identify the right targets for complex diseases, but also help us design appropriate molecules to treat and improve diseases while minimizing the impact of side effects.”

Together, he noted, these AI applications will help researchers put promising drug candidates into clinical development, with the ultimate goal of improving the likelihood of successfully bringing a drug to market and quickly getting new treatments to patients who need them most.

Many applications

In the field of new drug discovery, scientists are using artificial intelligence to make research faster, better, and more productive throughout the entire process:

Screening large libraries of drug compounds that are most effective against a specific disease.

Designing new drugs that are more targeted and less toxic than existing drugs.

Predicting side effects of drugs before they are tested in humans.

Individualize drug therapy for each patient.

As AI technology continues to evolve, we can expect more innovative and impactful applications of AI in research.