A Robot Scientist Might Have Found a New Malaria Drug

​Meet Eve, the "robot scientist." This bot may have just discovered a compound that could fight malaria.

UK researchers developed Eve as a tool to make the early stages of drug discovery faster and cheaper. Eve is a robotic system of inst​ruments capable of physically running tests—it includes a couple of robot arms, for instance—and also has a level of artificial intelligence. The researchers published a new paper on their work in the Royal Society journal Interface [TK link].

The researchers write that more efficient drug discovery "will enable the development of treatments for diseases currently neglected for economic reasons, such as tropical and orphan diseases, and generally increase the supply of new drugs." Orphan diseases are rare diseases that only affect a few people and therefore offer little financial incentive to develop treatments.

Discovering new drugs for diseases basically entails seeing if different compounds will treat a disease. It's a time consuming process that means conducting test after test. Eve can conduct 10,000 tests a day.

Eve doesn't just keep throwing molecules until something sticks. The paper explains that the robot starts by mass-screening a subset, then re-tests any that look promising in order to reduce false positives. It then starts with the ones that seem most probable and uses statistics and machine learning to figure out which compounds could be a winner.

"Such intelligent library screening may be more economical than standard mass screening as it potentially saves on time and compound use," the authors write.

To prove Eve was up to the job, the researchers gave the bot the target of dihydrofolate reductase (DHFR), a molecule from a malaria parasite that is in some cases b​ecoming resistant to existing drugs. It found that the compound TNP-470, which has been investigated in anti-cancer t​reatments, inhibits DHFR.

That's all very early stages in terms of drug development, and the process of developing it into a potential anti-malaria drug would involve a lot more work, but it demonstrates how Eve could help researchers on the way.

Human biochemists need not worry about being totally supplanted; the paper explains that this is a joint man-machine effort. A lot of the "man" part, however, gets pretty menial after they've defined the problem and engineered the testing procedures.

Other than that, their main jobs include maintaining Eve, replenishing stocks, and running different programs. The authors suggest, though, that the process could become more automated.

They argue that while later-stage drug development is the really costly part, automating the initial discovery phase could still have an economic impact.

"Preventing drug failures in late-stage development is an intrinsically very hard problem, as human biology is very complex," they write. "In contrast, we argue that a radical decrease in the cost and increase in the speed of drug discovery could be achieved by the full automation and standardization of procedures."