Key points from article :
Researchers recreated the performance achieved by DeepMind.
Unlike DeepMind, their method, which they dubbed RoseTTAFold, is freely available.
Since July, the program has been downloaded from GitHub by over 140 independent research teams.
“...developing RoseTTAFold for high accuracy protein structure prediction,” - David Baker, senior author.
It can reliably compute a protein structure in as little as ten minutes on a single gaming computer.
Used RoseTTAFold to compute new protein structures, including many poorly understood proteins from the human genome.
They also generated structures relevant to human health.
Can build models of complex biological assemblies in a fraction of the time.
RoseTTAFold is a “three-track” neural network.
Considers patterns in protein sequences, how a protein’s amino acids interact with one another, and a possible 3D structure.
Research by University of Washington School of Medicine published in the journal Science.