“What you just heard was tomato DNA moving through a nanopore,” David Eccles told the audience of TEDxWellington, after playing them the above track. The bioinformatician from New Zealand, who valiantly, albeit masochistically chose to study Biology because it was his scientific weak point in high school, has recently been working with genetic sequencing, specifically with the smallest and most cost-effective DNA sequencer around, the MinION.
As Eccles describes, the device’s speed means that in the future, doctors could run DNA tests to determine a patient’s illness, allowing for more specific diagnoses and treatments. Researchers have used the MinION to study the Ebola virus in Africa, and to study tree frogs in their natural habitat in the Tanzanian rainforest. The MinION has made all this possible within the last two years.
But, surrounded by an eclectic group of thinkers at TEDxWellington, Eccles was hit with new inspiration.
“I like to think outside the box. When I see wavy, squiggly things, I think of sound. Sound waves. What if we could listen to DNA?” He asked during his talk.
When DNA is sequenced using nanopore technology, the raw output is a file that looks like a soundwave. Eccles loaded the data collected from the live DNA sequencing into Audacity, then modulated its frequency to get the result we can hear now.
He was inspired to turn DNA into sound after hearing “Begin to Obsolescence,” the talk by fellow TEDxWellington speaker Asher, a.k.a. Skymning, whose lecture consists largely of a live set of analog synth music, and who introduced Eccles to making sounds in Audacity.
Unsurprisingly, the sequence sounds like an experimental modular synth song on fast-forward. Are DNA samples making their way to dancefloors anytime soon? Perhaps by an adventurous DJ. Eccles posted the file to Reddit’s Bioinformatics forum where he conceded, “If you slow it down about 10-40 times, it becomes a bit more musical.”
So that’s the sound of tomato DNA, but what about human DNA? Eccles tells The Creators Project that the two wouldn’t sound out of place on the same album. “DNA sequences are essentially random, but there are various physical and electrical constraints on DNA that mean that it will have a similar structure at the sequence level,” he explains.
“We also share quite a lot of our basic recipe for life. Both humans and tomatoes need to survive in the world–we breathe air, reproduce, and have energy demands, for example–so there are more than a few shared biological pathways.”
Besides the novelty of listening to DNA, a sound model can actually be helpful for researchers.
“We can't really look at DNA sequences yet, or at least not a single sequence on its own. All current sequencing technology works by converting the physical object of DNA into a model. Most typically, it's the 'four different letters' sequence model of DNA, and anything that doesn't agree with that model is swept under the carpet, or given fancy names like 'epigenetics.'
“The electrical model that Oxford Nanopore has developed is the closest so far that we have come to actually looking at DNA. I use sound as a proxy for vision because the electrical signal translates well into sound.”
A rough version of the talk is available online. Tune in at 1:29:00 on the video above to watch Eccles, and 5:26:15 for Skymning.