What does a Y-chromosome sound like? Now you can answer that question for yourself, using a novel molecule-to-melody conversion scheme that could open up new frontiers in biomedical research as well as computer-generated music.
Rie Takahashi and Jeffrey Miller of the University of California at Los Angeles describe the system in the open-access journal Genome Biology. They set up a system is to translate amino acids - the building blocks for human protein sequences - into musical chords.
This isn't the first time someone has tried to represent protein sequences musically, but Takahashi - an accomplished musician as well as a microbiologist - worked with Miller to come up with a more artful way to represent the standard 20 amino acids with the standard 13-note scale.
"The challenge was to find a way to be completely faithful to the science ... but also make the music more dimensional and add rhythm," Miller told me.
Takahashi added some extra twists: For example, similar amino acids are represented by the same chord - say, G-major for tyrosine and phenylalanine - but the arrangement is different. Also, more frequently encountered amino acids get longer notes than the less common ones.
With the aid of a colleague at UCLA, Frank Pettit, the researchers devised a Web-based program that can take the three-base code for each amino acid in a sequence, triplet by triplet, and turn it into a playable MIDI file.
Miller said the resulting music is completely determined by the protein sequence rather than tunefulness. "There are no fudge factors at all," he said.
The examples on the researchers' Gene2Music Web site range from horse hemoglobin to human thymidylate synthase A. "In principle, one could take the entire human genome and have it translated into 30,000 different protein sequences," Miller said.
Just for fun, I took a sequence from one of the markers on the human Y-chromosome and fed it through the converter. You can hear the MIDI result here - and it doesn't sound all that bad, if I say so myself. But you'll notice that there's a section where the same note sounds over and over. And that hints at the scientific application of the system.
The repeated notes are caused by repeating triplets in the protein code. In most cases, those repeats are harmless. But triplet repeats can also be associated with genetic neurodegenerative disorders such as Huntington's disease. Takahashi said one expert on Huntington's is already interested in using the musical system with his patients.
"It's a great way of explaining to the patient why the protein is dysfunctional," Takahashi said. "You can hear that through the repeated glutamines, over and over like a broken record."
The system could also be used to compare protein sequences by playing them together. A sharp-eared researcher should be able to hear the difference between a normal and abnormal protein. "What you'll hear is basically a dissonance or a difference in the chords," Takahashi said.
Miller said the technique might eventually be used to help vision-impaired researchers hear rather than see genetic code. "Admittedly, that's a future direction," he said.
The researchers' main goal is to use Gene2Music as a way to bring the joy of science to a generation raised on iPods and MP3 players.
"Music is a universal language and a bridge, and a way of making things interesting," Miller said. "For example, when I was a kid, 'Peter and the Wolf' was the way that young people got interested in classical music, because it had a different instrument for each character and it had a story. So that was a goal, to find ways to use music as a teaching tool."
I can imagine a day when getting your genome done will be as easy as getting your colors done - and when having a theme song based on your personal genetic code will be a musical status symbol.
Takahashi is already looking into tweaking some of the raw molecular melodies into polished musical compositions. "Ideally, I would like to complete a set of variations of different proteins, and ultimately make a CD of that," she said.