Robot scientist passes big bio test

Computers are always getting smarter. But sometimes the Watsons and Adams of the computer world generate big headlines. Now meet ABE (Automated Biology Explorer), a system that has generated mathematical models of complex biology from raw data, according to researchers involved in its development.

ABE, which is built from software developed at Cornell University called "Euriqa," was used to create models of glycolysis, a cellular energy-making process. To test the system, Euriqa was given data sets from measurements related to glycolic oscillations. With those raw data, the system churned out equations that were "nearly" the same as those understood to be correct. The group behind the study includes researchers from Cornell, Vanderbilt University and CFD Research Corporation in Huntsville, AL.

"What's really amazing is that it produced these equations a priori," Ravishankar Vallabhajosyula of CFD Research Corporation said in a statement. "The only thing the software knew in advance was addition, subtraction, multiplication and division."

ABE's feats have drawn comparisons to previous robot scientists. Adam, a system out of University of Wales at Aberystwyth, wowed the scientific community a couple of years ago with its new discoveries. Yet Adam's creators equipped the system with a model and access to a gene and protein database, ABE's developers pointed out. ABE generated equations with fewer human inputs than Adam did, they said.

The group's research can be found in the October issue of Physical Biology.

- here's the release

Suggested Articles

The new solution aims to streamline the incorporation of human genomic data into clinical trial designs.

TriNetX's platform uses EHR data to help drug developers with clinical trial protocol design and study site and participant identification.

The $58 million financing round represents biopharma industry's growing interest in genomics data.