Science Lab 2.0: Robot Scientists

0


A robot scientist working with teams at Aberystwyth, Cambridge and Manchester universities has demonstrated her potential capabilities in new drug discovery.

As quoted in a recent paper in the Royal Society journal Interface, and taken from a 2009 Science paper “A natural extension of the trend to ever-greater computer involvement in science is the concept of a robot scientist… A robot scientist automatically originates hypotheses to explain observations, devises experiments to test these hypotheses, physically runs the experiments by using laboratory robotics, interprets the results, and then repeats the cycle”.

eve manchester
Eve and a human colleague at Manchester University

Eve is designed to automate early-stage drug design. Her process starts by individually testing compounds from a large set, by brute-force method. The automation of this part of the process is not new or intelligent, and although Eve can get through 10,000 compounds a day, is still relatively time-consuming. With the added attribute of artificial intelligence, she improves upon this method by first executing experiments to retest and confirm each of the “hits” multiple times to eliminate false positives, which there can be a high rate of during the initial process. Analysing this list of confirmed positives using statistics and machine learning, Eve can then predict new structures that might perform better against the tests. She also exploits her intelligence to learn from early successes and select compounds to test that have a high probability of working against the chosen target.

Eve is not the first robot scientist, a title which belongs to Adam of Aberystwyth University, who started work in 2009, in the field of genetics. He worked on identifying which genes in microbes encode enzymes that catalyse reactions in yeast, and became the first machine to independently discover new scientific knowledge, albeit with a technical helper on hand, because both his body (hardware) and brain (software) are somewhat brittle.

Eve’s colleagues hope that her efficient process will help identify new candidates to tackle diseases such as malaria and other tropical diseases that are often neglected by the pharmaceutical industry because of the cost of drug discovery and the low economic return, even though they infect hundreds of millions, and kill millions, every year. Her career is definitely one to watch, and this may encourage others to employ more of her kind in the fight against diseases, alongside human co-workers.

avatar

Physics student and regular freelance science communicator, shooting for the stars. I’m your Science Editor and with the help of a team of talented writers, strive to bring you the most interesting and relevant science stories.

Leave A Reply