musical dna

dna

I am currently experimenting with sonifying the DNA strings of various plants in the garden.  Because of the repetition in the data the effect can sound quite musical.  As an example, here is what one of the gardens plants (Indian Bean Tree) sounds like:  LISTEN.

And here’s the DNA data: –

GAATCAATCCTACTACTTCTGGTTCNNNNGNNNCCACGCTTGAAAAAAAAAACC TGGGGCGTATCGTCCAAATCATCGGTCCGGTAMTAGATGTAGCCTTTCCGCCGG GCAAGATGCCTAATATTTATAACGCTCTAGTAGTTAAAGGTCGAGATACTGTTG GTCAACCAATTAATGTGACTTGTGAGGTACAGCAATTATTAGGAAATAATCGAG TTAGAGCTGTAGCTATGAGTGCTACAGATGGTCTGACGAGAGGAATGGAAGTGA TTAATACGGGAGCTCCTCTAAGTGTTCCAGTCGGTGGAGCGACTCTGGGACGAA TTTTCAACGTGCTTGGAGAGGCTGTTGATAATTTAGGTCCTGTAGATACTCGTA CAACATTTCCTATTCATCGATCTGCGCCTGCCTTTATACAGTTAGATACAAAAT TATCTATTTTTGAAACAGGAATTAAAGTAGTAGATCTTTTAGCCCCTTATCGCC GTGGGGGAAAAATCGGACTATTTGGGGGAGCTGGGGTTGGTAAAACGGTACTCA TTATGGAATTGATTAACAATATTGCYAAAGCYCAYGGGGGCGTATCCGTATTTG GCGGAGTGGGTGAACGTACTCGTGAAGGAAATGATCTTTACATGGAAATGAAAG AATCTGGAGTGATTAATGAAGAAAATATTGCAGAATCAAAAGTGGCTCTAGTTT ACGGCCAGATGAATGAACCGCCAGGAGCTCGTATGAGAGTTGGTTTGACTGCCC TAACGATGGCGGAATATTTCAGAGATGTTAATGAACAAGACGTACTTCTATTTA TCGACAATATCTTCCGTTTCGTCCAAGCAGGATCCGAAGTATCGGCCTTATTGG GTAGAATGCCTTCCGCTGTCGGTTATCAACCCACCCTGAGTACCGAAATGGGCT CTTTACAAGAAAGAATTACTTCTACCAAAGAAGGGTCCATAACTTCTATTCAAG CAGTTTATGTACCCGCGGACGATTTGACCGATCCTGCTCCTGCTACGACATTTG CACATTTAGATGCTACTACCGTACTATCAAGAGGATTAGCTGCCAAAGGTATCT ATCCAGCAGTAGATCCTTTAGATTCAACTTCAACCATGCTTCAACCTCGGATCG TTGGTGAGGAACATTATGAAACTGCGCAAAGAGTTAAGCAAACTTTACAACGTT ATAAAGAGCTTCAGGACATTATAGCTATCCTTGGGTTGGACGAATTATCCGAAG AGGATCGTTTAACCGTAGCAAGAGCGCGAAAAATCGAGCGTTTCTATCACAACC TTTTTTTGTAGCCGAAGTATTTACCGGTTCTCCAGGAAAATATGTTGGTCTGGC CGAAACCATTAGAGGGTTTCAATTGATCCTTTCCGGAGAATTAGATGGGCTTCC TGAACAGGCCTTTTATTTGGTAGGTAATATCGATGAAGCTACCGCGAAGGCTAT GAACTTAGAAATG

Advertisements

3 Responses to “musical dna”

  1. googleplx88 Says:

    Hi Robert,
    I was wondering why there seem to be a lot more notes than there are bases in the sequence you provide? Have you tried genes other than atpB?
    I have to admit that the generated noise does sound musical at times – but there are certainly better melodies that could be created with four (or more) notes.
    I’m looking forward to future iterations of the project.

    Cheers,

    g

  2. robert jarvis Says:

    Yes, you are right. Well spotted. The letters that you see are not what you hear. This is because I updated the audio to represent a fuller data set but never got round to updating the written-out bases. For this exercise I also paid little attention to how it sounded, i.e., what notes were chosen, and so you are also right: it could sound more musical.

    For me the concept became more interesting after I had worked out how to map the base data to the Amino Acids. There are 18 of these and so by a careful choosing of the notes, in my opinion, more interesting melodies could be heard. Please see the “Acid Music” post above for a description of how I did this. There is also a musical example in the post (at:
    http://www.robertjarvis.co.uk/sound/thylakoid.mid).

    Once I had the data, I allowed myself more artistic freedom in order to pull out the musicality of my various genetic starting points. Maybe I should post some final examples up on the site, or you could always hear it in situ, as it will be featured again in the garden this year – between May and October.

  3. googleplx88 Says:

    Hi Robert,

    That is fascinating – I must admit that I had a similar thought concerning the use of the amino-acids in the protein sequence as the basis for tone generation.

    I see that you have used the molecular weight of the aa’s to determine the tone of the note; I was thinking about a different direction which may result in a more musical (and informative) presentation:
    I was thinking that the character of the aa could determine the associated tone. For instance, assigning notes of a particular scale to hydrophobic residues. This would let you “hear” trans-membrane domains if hydrophobic aa’s were keyed to, for instance, a minor key. I was also thinking that proline would serve well as a pause or “empty note” as that residue frequently acts as a “helix breaker” and an end to trans-membrane domains, which would create a pause between domain (and key) changes. Residues often found in hydrophilic alpha-helices could also constitute a population of aa’s which could represent a tonal color. Lastly, infrequently used charged residues such as lysine could serve as “out” notes for the particular scale chosen which would add a burst of tonal color when these residues appear.

    As I’m sure you have deduced by now, I work in the molecular biology field and am a music lover. I would like to reiterate how fascinated I am by your work, and how cool it has been to find a project which is an amalgam of two things that are very dear to me. This is a wonderful merging of art and science, and I would love to see it “proven” that music is in my soul.

    Cheers,

    g

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: