to Nature home page
home
search






Nature15 January 2004

 nature highlights

Automated research: Doctor Robot

In many areas of science, data are being generated faster than they can be effectively analysed. So the question of whether the scientific process can be automated has become of more than theoretical interest. King et al. present a proof-of-principle experiment that shows that a robot scientist is a realistic concept, at least in functional genomics. The robot was built using lessons learned in studies on artificial intelligence. It can automatically originate factual hypotheses to explain data, devise experiments to test the hypotheses and instruct a lab robot to perform the experiment. It then assesses the results and runs more experiments if required. The robot's first work was to determine gene function using yeast deletion mutants and growth experiments.

letters to nature
Functional genomic hypothesis generation and experimentation by a robot scientist
ROSS D. KING, KENNETH E. WHELAN, FFION M. JONES, PHILIP G. K. REISER, CHRISTOPHER H. BRYANT, STEPHEN H. MUGGLETON, DOUGLAS B. KELL & STEPHEN G. OLIVER
Nature 427, 247–252 (2004); doi:10.1038/nature02236
| First Paragraph | Full Text (HTML / PDF) |

15 January 2004 table of contents

  
  © 2004 Nature Publishing Group