Contrary to first impressions, an automated system that designs its own experiments will benefit young molecular geneticists.
At first glance, it seems to render obsolete the armies of postgrads and postdocs employed in the world's molecular-genetics laboratories. In this week's issue (see page 247), a British team unveils an automated system that “originates hypotheses to explain observations, devises experiments to test these hypotheses, physically runs the experiments using a laboratory robot, interprets the results to falsify hypotheses inconsistent with the data, and then repeats the cycle”.
What's more, when set loose on experiments to investigate the genetic control of important metabolic pathways in yeast, it performs more cost effectively than scientifically educated human volunteers. The Robot Scientist seems to promise a future of successfully completed research projects, untouched by human mind.
Neo-luddites must be unsure whether to curse or celebrate. On one hand, they are obliged to condemn another technology that seems to threaten established patterns of employment. But they may also be glad to see the scientific and technological élite seemingly hoist with its own petard.
The truth is rather different. The Robot Scientist does represent an important step forward, but does not spell the end for its human counterpart. The deductive steps required to design experiments for functional-genomic analyses are particularly amenable to solution by computer algorithms. And this is a field in which the deluge of data requiring explanation exceeds researchers' capacity to cope.
The team behind the Robot Scientist argues that such automation “frees scientists to make the high-level creative leaps at which they excel”. Therein lies the challenge. Some lab heads still treat postgrads and postdocs as a cheap source of menial labour, rather than educating them to become tomorrow's creative research leaders. We can only hope that the Robot Scientist helps to change such attitudes.
About this article
Cite this article
Don't fear the Robot Scientist. Nature 427, 181 (2004). https://doi.org/10.1038/427181b