Saturday, March 14, 2026

Where are all AI drugs?

Share

Fresh medicine It usually starts with a tragedy.

Peter Ray knows that. Born in Zimbabwe, a child’s mechanic and radiology technique, Ray escaped with his family to South Africa during the Liberation War in Zimbabwe. He remembers the journey in 1980 in the convoy of armored cars. When the sun was burning, the soldier taught an 8-year-old radius on how to launch a machine gun. But his mother still had to stop. She didn’t feel good.

Doctors in Cape Town diagnosed her cancer. Ray remembers how radiotal treatments with her, hospital rooms, and colostomic bags. She loved the beach, she loved to walk on the line in which the water met the earth. But it was more arduous to go. Sometimes she returned home from the hospital for a moment and it seemed that everything would be better. Ray raised his hopes. Then everything will fall apart again. Surgery, radiation, chemotherapy – treatments that were on the table in the 1980s – were soon exhausted. When she lay dying, he promised her that he would somehow make a difference. He was 13 years aged.

Ray studied to become a healing chemist, first in South Africa, taking loans to finance studies, and then at the University of Liverpool. He worked at pharmaceutical companies in Great Britain, with many projects. Now, at the age of 53, he is one of the main drug designers in a pharmaceutical company called Recursion. He thinks a lot about this promise for his mother. “He lived with me all his life,” he says. “I have to get medicines on the market that affect cancer.”

The desire to stop your own tragedies from someone else can be a mighty motivator. But the drug discovery process has always been grinding, exhaustively snail-paced. First of all, chemists such as ray zero on their purpose – usually protein, a long string of amino acids rolled up and on themselves. They call on the model on the computer screen and watch how it turns into a black void. They notice curves and declarations on its surface, the places where the molecule, goodbye to the darkness like a spacecraft, could have anyway. Then, atom atom, they try to build a spaceship.

Animation: Balarama Heller

When the modern molecule is ready, chemists give it to biologists who test it on live cells in balmy rooms. More tragedies: many cells die for reasons that are not always clear. Biology is convoluted and the modern drug does not work as expected. Chemists will have to create another and the other, improving, adapting, often for years. One of the biologists, Keith Mikule from Insilico Medicine, told me about his experience in another pharmaceutical company. After five years of work, their best molecule had unforeseen, perilous side effects, which meant that they could not go further. “There was a large team of chemists, a large team of biologists, thousands of molecules and there is no real progress,” he said.

If the band is lucky, they get a molecule that does what should do with mice. They have a chance to convey a miniature group of vigorous human volunteers, phase I study. If volunteers retain health, it gives more people, including people from a given disease, in phase II. If the patients do not get diseased, they have a chance – phase III – to give it a larger number of patients as many as possible, as possible, as diverse as possible.

At every stage, for the reasons why few people understand, and less can predict, great drug rafts decrease. Along the way, over 90 percent of hope fails. When you meet drug hunters, you can ask them carefully, tenderly if they ever had a drug. “This is very rare,” says Mikule, who has one medicine (nirarib, for ovarian cancer). “We are unicorns.”

Latest Posts

More News