Conversion of amino-acid sequence in proteins to classical music: search for auditory patterns
© BioMed Central Ltd 2007
Published: 3 May 2007
Skip to main content
© BioMed Central Ltd 2007
Published: 3 May 2007
We have converted genome-encoded protein sequences into musical notes to reveal auditory patterns without compromising musicality. We derived a reduced range of 13 base notes by pairing similar amino acids and distinguishing them using variations of three-note chords and codon distribution to dictate rhythm. The conversion will help make genomic coding sequences more approachable for the general public, young children, and vision-impaired scientists.
In an effort to make science appealing to a wider audience, interdisciplinary groups have combined efforts to initiate novel approaches for the presentation and perspective of scientific material. An example is that of Victor Wong, a blind meteorology graduate student studying at Cornell University. He developed a computer program that translates different colors of a weather map into 88 distinct piano notes. With the use of a stylus to scan across a weather map, Wong was able to hear a gradation of colors ranging from blue to red with respect to electron density . Another example of an interdisciplinary approach involves Japanese biologists at the RIKEN Center for Developmental Biology in Kobe, who have incorporated basic concepts of developmental biology into card games based on manga characters like Pokémon to interest young people. Aside from the amusing, colorful characters, the creators hope to preempt the introverted, asocial stereotypes of scientists before they "take root" . Also, the Biochemist's Songbook by Harold Baum describes scientific concepts with lyrics and song .
In the context of basic research, a conversion from genomic sequences to music would open a door for the vision-impaired to study genomic biology. An auditory presentation could also be a means of exposing students to the concepts of DNA sequences and protein sequences at an earlier age through the use of auditory characteristics such as length, tempo, and dynamics. Some studies have attempted to transpose DNA sequences directly to music . This approach suffers from a limited number of notes based on nucleotides composed of only four bases: adenine (A), cytosine (C), guanine (G), and thymine (T). Although the DNA could be read as a note for every two or three consecutive bases, this would focus the melodies more on DNA sequence organization and be less informative than looking at the coded information per se. Moreover, the result creates a string of notes that has no recognizable theme or musical depth as a composition. Other attempts to convert DNA sequences to music have used mathematical derivations based on the physical properties of the individual nucleotides in codons to create a set of equations for translating DNA sequences to musical notes [5, 6]. A number of studies have dealt with pure protein sequences [7–10]. For example, Dunn and Clark used algorithms and the folding patterns of proteins to translate amino-acid sequences into musical themes . Such an assignment creates a range that spans two to four octaves. Notes spanning such large ranges typically yield scores that lacked musicality. They also examined a nine-note scale, but without distinguishing among amino acids having the same note value .
The goal of our work is to find a mode of converting genomic sequences (including coding and, eventually, non-coding) to piano notes that sound reasonable to a musician's ear while remaining faithful to the science of the protein sequences. The classic problem to overcome is the jump between consecutive notes as a consequence of the 20-note range when each amino acid is represented by a unique note. The wide range of the notes results in melodies that have many large, sporadic jumps, making them difficult to follow musically. A second problem is the question of how to incorporate rhythm into the sequence of notes. We describe here several innovations in coding assignments that generate a reduced note range and that also introduce rhythm into the sequence of notes.
By converting genomic sequences into music, we hope to achieve several goals, which include investigating sequences by the vision impaired. Another aim is to attract young people into molecular genetics by using the multidisciplinary approach of fusing music and science. There are strong associations between music and perception. Heightened interest in a historically known condition called synesthesia (or synaesthesia) has also spanned multiple fields of study including science, music, and history . The condition has prompted a collaborative approach among various disciplines aimed at developing a more comprehensive picture of this syndrome. Synesthesia is an involuntary perception produced by stimulation of another sense. Commonly one hears a certain pitch that consistently evokes a particular color. Synesthesia is considered an unusually strong cross-modal association in the brain and has been observed in children and adults . Another example of a collaborative, cross-disciplinary effort includes research pertaining to sound-induced photisms. Sound-induced photisms have been recorded where a startled reaction to a sound (soft or loud) evokes colors ranging from flashes of white light to a colorful flame . A separate study confirms that lighter colors 'fit together' with higher pitches of sound and darker stimuli are better fitted to lower pitches .
In future studies, we will use a recently created program (F. Pettit, unpublished work), now in its testing stages, which implements the translation rules we have formulated. Use of this program will enable very rapid translation of large segments of genomes into music. Furthermore, different instruments can be assigned to unique parts of the genome, such as regulatory, intergenic, and promoter/operator sequences, in order to use the obvious distinction as a teaching tool for introducing the function of the genome and its parts. Finally, each protein provides a theme that can be used as a source to make variations that would involve improvisation and elaboration, which would allow the investigator/author to contribute an artistic component to the original melody. For further examples of protein music and references to previous work, go to our website gene2music . Also, browse this website to access our computer program in order to convert your own gene of interest to music.
The following additional data are available with the online version of this paper. Additional data file 1 is a music clip of the human ThyA protein based on the single note assignment of one amino acid per musical note. Additional data file 2 is a music clip of the human ThyA protein derived from the reduced 13-base note chord assignment. Additional data file 3 is a music clip of the human ThyA protein based on our final coding assignment, which includes rhythm. Additional data file 4 is a music clip of the huntingtin protein based on our final coding assignment.