Music in the genes...laah-lah-laah-lah-lah....
(June 4th 2007) Do DNA and protein sequences sing unsung songs? That seems to be a slightly far-fetched idea. Yet, some researchers are converting those sequences into melodies. The newest contribution was created by Rie Takahashi and Jeffrey Miller. Is it the same old song or a new chart breaker? A query from Anja Possart.
It's no secret that scientists behave a bit cranky from time to time. Or how else would you regard scientists converting genomic and protein sequences into music? Are they doing it just for fun, are they searching for a deeper aesthetics of scientific data? Or do they simply want to make science more appealing to a wider audience?
Whatever the reason, there have been umpteen approaches to this topic. Just to mention some of them, for instance, nucleotides were arranged according to molecular weight and they were transformed into musical scores based on an eight note scale. In another attempt, DNA was exposed to infrared light and the measured ratios of absorbed light frequencies were converted into ratios of sound. Alternatively, one of four notes spanning a five-note interval range was assigned to the four DNA bases. The range was chosen to be a fifth of the human voice range and the notes were chosen from the middle of the C Major scale for reasons of symmetry.
Sounds elaborate ... at least to those who, like myself, only understand the half of it! However, the musicians among the life scientists aren't satisfied with the outcome. Rie Takahashi and Jeffrey Miller from the University of California, Los Angeles, USA, recognised the lack of musical depth of nucleotide and amino acid compositions written so far and developed a new mode of converting genomic sequences to piano notes that "sound reasonable to a musician's ear while remaining faithful to the science of protein sequences" (
Genome Biology 2007).
Producing a reasonable sound certainly is the more difficult task; just consider for a moment the long rows of sequence letters many scientists stare at each day. Where in all these A's, C's, G's and T's or within all the Ala's to Thr's would you expect musicality, except for the monotonous humming of your computer? Music would be a joyful variation during a working day.
The direct transformation of DNA sequences to music definitely suffers from a limited number of notes, based on the fact that there are only four nucleotides. However, this didn't keep biologists and composers from assigning tones to each nucleotide and using them as base lines to accompany several melodies. The resulting music might be rather "reasonable". But where is the hidden scientific meaning?
Let's try proteins. Considering musical depth, the use of twenty components is much more promising. But, on a closer look, this just leads to the opposite extreme. When each amino acid is represented by a unique note, scientists suddenly face large jumps in pitch that make it difficult to follow the melody - not to mention rhythm and harmony.
Takahashi and Miller solve these problems as follows:
In order to create a fuller sound, the biologists substitute amino acids not by single notes but by triads. The large range of almost three octaves is reduced by pairing similar amino acids such as tyrosine and phenylalanine. To keep the scientific information, the members of each pair can be differentiated using a root position triad or first inversion triad. Furthermore, the authors sectioned the codon distribution in the human genome into four parts, each of which is represented by either an eighth, quarter, half or whole note. The more abundant the codon, the longer the associated note duration. This is about as far as laymen can follow the theory. Anyway, the result is what counts. You will find the music clip of the human thymidylate synthase A (Thy A) and other proteins together with the online version of the paper, on the scientists' website (
http://www.mimg.ucla.edu/faculty/miller_jh/gene2music/home.html).
A programme which has recently been created in collaboration with Frank Pettit, University of California, also enables the translation of any given protein sequence. By providing that tool for rapid conversion of genomic sequences into music, the authors hope to make molecular biology more accessible to visually impaired scientists and to the general public.
Although I have tried to enjoy the "Thy A" tune, I still have to conclude that, to my ears, Mozart displayed a bit more sense of harmony. To overcome my philistine ignorance, maybe I should try to enthral the audience at my next work progress report by allowing "my" protein to play the fiddle.
