Communications of the ACM
Test Tube Artificial Neural Network Recognizes 'Molecular Handwriting'
Researchers at the California Institute of Technology have developed an artificial neural network from DNA that demonstrates the capacity to program artificial intelligence into synthetic biomolecular circuits.
Credit: Olivier Wyart
California Institute of Technology (CalTech) researchers have developed an artificial neural network from DNA that can correctly identify handwritten numbers, demonstrating the capacity to program artificial intelligence into synthetic biomolecular circuits.
The team plans to program intelligent behaviors—such as the ability to compute and make choices—into artificial DNA-based neural networks.
The researchers showed that a neural network comprised of carefully designed DNA sequences could carry out prescribed chemical reactions to accurately identify "molecular handwriting," with each "number" comprised of 20 unique DNA strands chosen from 100 molecules, each assigned to represent an individual pixel in any 10-by-10 pattern.
When presented with a particular example of molecular handwriting, the DNA neural network can classify it into up to nine categories, each representing one of the nine possible handwritten digits between 1 and 9.
From Caltech News
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found