Communications of the ACM
Machines Whisper Our Secrets
The sounds of laboratory equipment in use can be utilized to reconstruct what they were doing, researchers say.
Credit: Darryl Leja and Ernesto Del Aguila III, National Human Genome Research Institute
Researchers at the University of California, Irvine (UCI) and the University of California, Riverside have developed a method for reverse-engineering how laboratory instruments are used, based on the sounds they produce.
The researchers tested their hypothesis to differentiate the sounds made by a DNA synthesizer as it manufactured specific sequences of adenine (A), guanine (G), cytosine (C), and thymine (T), the four constituent bases of nucleic acids.
Said UCI's Sina Faezi, "Through a careful feature engineering and bespoke machine learning algorithm written in our lab, we were able to pinpoint those differences," and easily distinguish each time the machine generated A, G, C, or T.
When the researchers used software to analyze the AGCT patterns they obtained through the recordings, they identified the correct type of DNA with 86% accuracy, then boosted accuracy to nearly 100% by running it through additional DNA sequencing software.
From UC Riverside News (CA)
View Full Article
Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA
No entries found