Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Music Media Science

Protein Music 84

Posted by michael from the adenosine-sharp dept.
isudoru writes: "Ever wondered what the DNA can be converted into? Well the first attempt to convert the letters of DNA is to make them into music. A program called ProteinMusic converts the DNA sequences into midi songs and is able to export to midi too. It's platform independent since it's made in java."
This discussion has been archived. No new comments can be posted.

Protein Music More

Protein Music

Comments Filter:

  • JVM not a platform (Score:1)

    by Anonymous Coward
    pedantic point, but isn't a compliant JVM a requirement of running any Java app? it's not really platform independent...more OS independent. the plaform here is the JVM...

  • The sad part is that (Score:2)

    by Anonymous Coward
    these seemingly random notes sound better than the crap you hear on the radio nowadays.
  • DNA music was first created 3 years before Children of God was released.
    Not that it isn't a fairly good book although The Sparrow was much better, even though its characters were horribly stupid in a Star Trek sense.
    "Ooh. After a journey of many years over light years of space, we have arrived at an alien planet we know nothing about." Welp, atmosphere seems earthlike, lets land and see if there are any toxins on the ground that will kill us. Or maybe some man-eating flowers."
  • Companies that have already patended certain DNA sequences could sue you for divulging trade secrets. A patent (properly) discloses the inner workings of an invention. In theory, anyone can, after the monopolistic term has expired, can create a duplicate of the original invention, simply by following the instructions laid out in the patent. This is quite different from trade secrets, which, naturally, involve no public disclosure.
  • The more I have thought about it, the more I wondered if any 4 notes, if cleverly chosen, will sound musical if played at random.

    Actually, the version of the DNA music program shown here first translates the sequence to amino acids, and so has 20 notes.
  • Nits are welcome! My point was that there's much room for creativity, even within a rigid structure. The folks who try to make "music" from DNA, as opposed to "auditory information" from DNA, are doing exactly that -- being creative within a certain framework.

    It's a little like fractals (but not quite) -- all those fancy colors don't exist at all in the Mandelbrot set. They don't cycle, they're not all trippy (except, of course, to mathematicians). The visual representation is one way of displaying the information. The equations just aren't enough for most folks. But given that people work more off their eyes than their ears, the value of converting information to sound is yet to be proven.

  • I remember seeing this back in the 70's. Some TV science show, and they went to the trouble of masking out all but four keys on a piano for the demonstration of the "music of life". And, I'm pretty damned sure there's someone who's put out an album of DNA "songs". People commented how "beautiful" the "music" sounded.

    If people want to fawn about how "beautiful" the "music" is, they're missing the point. Humans are, if nothing else, pattern-seeking computers. There isn't software yet that can spot trends in art, obscure connections between films, influences in music, etc.

    Transforming information into "music" in a predictable way is just another way of receiving the information and giving us the chance to process it . If we weren't such visual creatures in the first place (imagine if Bat Boy [batboy-themusical.com] were the rule rather than the exception), we'd be doing much more of this as a matter of routine. But as we trust our eyes more than we trust our ears, we tend to think of these applications as novelties. The fact that we see music as mere "entertainment" rather than as information makes it even more difficult for folks to see the real value here.

    For this to be meaningful as a source of information, or a different way of looking at information (is there a difference?) then there should be no choice of duration, octave, instrumentation, harmony, etc. unless it makes hearing "things" easier, or if it is determined from the sequence. (every time you see GTACGG, shout "YIPPIE!") But if the performer gets to choose, we're hearing the performer's composition based on the information rather than a straightforward presentation of the information. Take a look into 12-tone music [artswire.org], for example.

    Maybe after years of work on such things, we'll develop some decent guidelines on how to present information with sound, and Edward Tufte will come out with The Audio Presentation of Quantitative Information or Enlistening Information.

  • And just a little more time before there are people suing them :)
  • wondering how a JAva program is going to get my DNA sequence, especially since I don't have a DNA sequence finder setting around in my room?

    You DNA sequence is almost identical to that of every other human being. Over the size of the genome, the differences amount to almost nothing - in terms of the amount of difference it would make to a piece of music, you're unlikely to notice any difference between different individuals.
  • ...if you don't want to put SlowJava on your pc.

    Try Java. It has improved over the last five years. Don't dismiss it just because it was slow when it first came out.
  • But JVMs are available for almost all OSs. Also, you don't need a JVM to run a Java program. You could also compile it to an executable using a native compiler.
  • The idea of encoding information as music isn't a new one. Unfortunately most attempts don't attempt to convey any of the original meaning in the new music.

    Here's DeCSS as music: http://www.cs.cmu.edu/~dst/DeCSS/Gallery/Castleman /css_descramble.mid [cmu.edu].

  • Convert that summary into english?

  • This was started back in 1995 (Score:5)

    by alteridem ( 46954 ) on Sunday June 17, 2001 @05:51AM (#145680) Homepage
    A UK band called the Shamen started this back in 1995. The song S2 Translation [cuttlefish.com] on the '95 album, Axis Mutatis [cuttlefish.com] was generated using the software developed by one of the band members, Colin Angus. Colin worked with R. King who is now developing the Java version. The song was formed by translating part of the human DNA sequence of the S2 protien. The software was written for the Mac in C and as far as I can tell isn't available anymore. I only had a PC back in 95, so it is nice to see them still working on this and porting it to Java so we can all try it out.

    If you want to check out a sound sample from the song, see this page [cuttlefish.com].

  • This reminds me of pi music [escape.com]. Isn't nature beautiful?
  • If I write 23 bars of DNA music and my girlfriend writes 23 bars of DNA music, am I going to be paying royalties to her for the rest of my life?

    Only if you make beautiful music together :)
  • would prefer more advanced stuff, algorithmic composition, aritficial intelligence and electronic music.

    The linked article doesn't cover how the DNA pattern is represented as sound, how the sequence is mapped to a set of pitches. Given the limited combination of four values, my knee-jerk reaction is that a straight representation of a sequence would be musically boring.

    But then, it's not really supposed to be music, it's supposed to be a graphic representation, a chart.

    I think it would be more cool if someone dug up some info on holographic audio representation of earthquake phenomena.
  • ...opening night at the DNA Lounge [dnalounge.com].

    I can imagine the kind of patter you'll be hearing on the dance floor...

    "Yeah- its got a good beat, you can catalyze to it. I give it a 9."

    "I just wish they'd have considered adding a RNA polymerase, you know? Its missing something. You ain't gonna be synthesizing mRNA if you can't transcribe the nucleotide sequence."

    "True. True..."

  • Have you wondered what you get if you sequence and play the protein thought to be responisble for Alzheimers?

    Brittany Spears, NKOTB, and N'Sync.

  • Feel DNA models? Time for a girlfriend, eh?

  • You heard it here first!
  • Dunno, I'd put the protein-folding in the "run-time behavior" category, not the compilation phase.
  • I guess I'd better qualify my statement - I'd put the "default" folded state for the molecule as a compilation.

    The dynamic folding behavior of the molecule in the context of an active environment would be equivalent to its execution (perhaps as a thread?).

    Its interactions with other molecules would be my equivalent to inter-process/thread communication.

  • That might work for protein structure, but when we're talking about straight DNA samples, there's not always a correlation between the DNA sequence and the amino acid chain. Not every start codon is interpreted, and sometimes it's on the other chain of the DNA double helix. In other words, your song might be backwards and wrong:)

    Must admit, it's alot more interesting sounding than music with four tones. THough if it's four chords and played on a guitar, then it's more inventive than most modern rock music..

  • My great-grandfather had this idea in the 50s. He was always talking about creating music from DNA, but we just brushed him off, teasing him all the time. At that time, it seemed impossible. Now we know better. I wish he had made a patent on it just before he died. Surely, nobody had come up with this idea prior to him, so we should be the ones to cash in on this.

    Moderators: It's +1 s a r c a s m.

    - Steeltoe

  • Just a matter of time... (Score:5)

    by tomson ( 100060 ) on Sunday June 17, 2001 @04:13AM (#145693)
    before DNApster.com starts it's service :)
  • "from the adenosine-sharp dept."

    Yes, adenosine is a protein, but I think you meant adenine, since the 4 base-pair proteines are adenine, thymine, cytosine, and guanine.

    -Josh
  • Wow, then he was even more wrong than I thought.
  • Surely if it sounds like trash then it's your parent's fault?

    Unless, of course, you've got some ST:Voyager-style retro-viruses that have been re-writing the O-Town tunes in your base-pairs as Bach or Rob Zombie whilst removing those nasty Borg implants etc.

  • Re:This was started back in 1995 (Score:3)

    by zulux ( 112259 ) on Sunday June 17, 2001 @06:50AM (#145697) Homepage Journal
    Actually the idea was floating around before that - I did a small program in Turbo Pascal that tweaked the rather small encoding of a virus strain into *horible* music. I found the sequence in a 1987 issue of Scientific American and coded my program in 1990.

    I don't remember where I got the idea from - but I know I got it from sombody else in the 1989 time frame.

    Makeing tones on the PC-buzzer using the viral code was easy - coaxing them into music was hard. Turbo Pascal came in handy because I could tweek the code recompile quickly.

    I remember how hard it was for me to find out the diference in Hz between notes and what Hz "middle c" started at- this was before the internet made it's way into my life and I had no musical training.

  • well sorry but dna isn't converted into anything

    it's not source code, it's parameters
    .oO0Oo.

  • Since this article has "music" in the title, lots of people have brought up the RIAA and copyright law, asking things like "who owns the music?", "will I have to pay royalties on DNA?" and "if it's an NSync song, can I be sued?" IANAL, but I have a pretty good grasp on the state of the law.

    Just because the DNA came from your body doesn't mean the music is automatically yours. See Moore v. Regents of the University of California, which held that a development based on a removed spleen is not the property of the splenectomy patient.

    Eben Moglen points out in "Anarchism Triumphant" that all information can be reduced to numbers; from there it's obvious that it can be represented in base four or base twenty. While Moglen tries to use this to attack IP law, asking how one number can be copyrighted and another patented and another free for use, others have correctly pointed out that what's being protected is not the number, but the number plus the translation scheme. In this case, DNA is not copyrightable without some additional work of authorship,

    Copyright protects only against copying, not against independent creation. If someone's DNA happened to sound like a copyrighted song, a plaintiff would pretty much have to show that the translation scheme was created with the intent to infringe. DNA altered to produce a copyrighted tune would also infringe. Ordinary DNA with a previously developed translation scheme would not infringe, because it would be obvious that any overlap was pure coincidence. (However, calling attention to that overlap in the market would probably be trademark infringement.)

  • What are you talking about??

    As for how you are going to come up with input to this program, let me help you out:
    Rip all the keys off of your keyboard except for A, T, G, and C. Now start typing. It's not that complicated...
  • I am sure I could replicate the exact same effect by applying sandpaper to my genitals.
  • More to the point, will the RIAA be suing us all because we are carrying copyrighted music around inside our bodies ?

    Will pregnant mothers be sued under the DMCA for illegally reverse engineering encrypted music ?

  • Well, just having one's DNA translate to an N'Sync song would itself me the most terrifying thing one can imagine.

    Although it is more likely that a virus, not a human, would have DNA that translated to that.

  • It IS source code. Proteins are the object code. Of course, the compiling algorithm is very simple. Every 3 bases (4 different bases - i.e. 2 bits a base) code for one amino acid. 3*2=6 2**6=64 and there are less than that many amino acids, some are aliases of others, some (3 of them I think) stop protein coding, and some (I think 1 one them) starts protein coding.

    So maybe it is more like assembly language source code.

  • Writing something in Java does not necessarily mean that it will run on any platform. At a minimum it means that it will crash on any platform.

    Having said that, I'm going to relive my days as a DNA mapping software author and download this program.

  • made this: Molecular music [molecularmusic.com] Nice thingy )

  • The Shamen did this in 1995 (Score:3)

    by pesc ( 147035 ) on Sunday June 17, 2001 @04:31AM (#145707)
    I have the recod Axis Mutatis by the Shamen.

    amazon link [amazon.com]

    The last song, "S2 translation" has the following text on the sleeve:

    "sequenced from the amino acid characteristics and the DNA coding for protein S2, receptor for serotonin and other tryptamines, thanks to Dr R.D.King for the data."

    As you can see in the referenced article, Shamed and R.D.King worked on this a long time ago.

  • Well, the DNA is converted to RNA which then does the actual amino acid choosing.

    There are only 19 amino acids that make up proteins.
    The RNA seqence to start a protien is AUG and the stop sequences are UAA, UAG, or UGA.

    --
  • All I can get it to play is Taps!!! *shudder*
  • ...TTCCGATACGATTAGCC da da doo doo bop bop wobba do-day hubba ma moo ya ya ba BOOEY!
  • Since Java isn't open source, there's no telling what could happen to your DNA! Sun has code hidden within Java to steal your genetic information and email it to Microsoft!

  • Ratios and Mozart, etc (Score:3)

    by Alien54 ( 180860 ) on Sunday June 17, 2001 @05:07AM (#145712) Journal
    I remember someone did some statistical analysis on something just like this many years ago, as far as distribution of frequencies etc. The end result was that the distribution curve in the DNA matched that of large pieces of music, such as by Mozart, etc.

    I wonder about 1) how albums by certain rock artists would match up in this regard; and 2) the algorthm for the original conversion of DNA to music. There are so many ways that it could be done. You could take the 21 amino acids and line them up on any number of scales: chromatic, wholetone, diatonic, etc. Some choices would generate more musical results than others.

    Check out the Vinny the Vampire [eplugz.com] comic strip

  • Back in the 80's there was some kind of DNA music program available for the Amiga. My aged memory cells do not provide me with the program name, but if no one here remembers (or believes), I'll try to track it down.
  • What I'm interested in is, what if you took a Mozart piece, converted it back into DNA, and spliced that DNA into a test tube baby? Would the child be a genetic mutation that is horrid to look at, or a perfect human?
  • It also brings a whole new meaning to "Faire la Java".

    (Note: for non-french speakers, this means "to party")

  • This idea seems to come straight out of the SF novel "Children of God" [amazon.com], the sequel to the should-be-classic "The Sparrow" [amazon.com] by Mary Doria Russell. I never imagined that converting DNA to music would ever actually prove practical, though.
  • I saw this done on TV news a long time ago, before the 1996 date quoted on the site. The implication then was, "Music comes from our genes." I thought it was pretty cool at the time, so I was surprised that years passed without hearing about it again. The more I have thought about it, the more I wondered if any 4 notes, if cleverly chosen, will sound musical if played at random.

  • Here's something DNA can be converted to... (Score:3)

    by empesey ( 207806 ) on Sunday June 17, 2001 @03:49AM (#145718) Homepage
    Babies.

    Who would have thunk it?

    --
    Sometimes the obvious answer is the correct one.

  • Legal Questions (Score:4)

    by empesey ( 207806 ) on Sunday June 17, 2001 @05:10AM (#145719) Homepage
    If I write 23 bars of DNA music and my girlfriend writes 23 bars of DNA music, am I going to be paying royalties to her for the rest of my life?

    Who's going to get custody of the gold records that hang on the wall?

  • Is there some way to convert a song to a DNA? Could I reconvert all my MP3 to DNA sequences? If this could be possible... when will be released a DNA player for my car? ;)
  • ok - so if I create music that is based on the portion of my DNA that creates my penis will it sound anything like:

    boom chika boom bowow bum boom chika boom bowow?

    just curious...

  • This gives a whole new meaning to his SF novel "Blood Music" (btw, it was about single-celled intelligent organisms... sure their DNA must be more complicated than a Beethoven score :-))
  • no shit...i mentioned that...you might want to read the rest of my post....

  • and how is the program going to get my DNA? (Score:3)

    by unformed ( 225214 ) on Sunday June 17, 2001 @03:55AM (#145724)
    wondering how a JAva program is going to get my DNA sequence, especially since I don't have a DNA sequence finder setting around in my room?

    However, the implications of this are pretty cool....(could be gross violation of privacy too)

    once they can get everyone's DNA sequence (not even all of it, since 99% is the same, but just the unique portion), and then convert it to a number, it can serve as a unique encryption key/id whatever...although it could also kill anonymity...

    hmmmm....
  • The compiling algorithm is not simple! It is very hard. That is called protein folding. It is an unsolved problem. See the IBM Blue Gene project.
  • Actually it is a pretty cool tune. S2 Translation that is.
  • "run-time behavior" would be folded proteins interacting with other molecules.
  • I'd put the "default" folded state for the molecule as a compilation.

    Ok.

    The dynamic folding behavior of the molecule in the context of an active environment would be equivalent to its execution (perhaps as a thread?).
    Its interactions with other molecules would be my equivalent to inter-process/thread communication.

    I think this is getting too fine-grained and perhaps making distinctions that don't really exist and it's better to leave it as the idea of a folded protein interacting with its environment as its "run-time behavior."

  • I'll do you one better...

    How long before DNA gets labelled as a copyright-circumvention device?

  • Adenosine is not a protein. It is a nucleoside (base + sugar). Add phosphate group(s), and you would have a nucleotide.
  • I used to be a big fan of protein and DNA until they got big and sold out.

  • Think of the confusion (Score:3)

    by Millyways ( 262662 ) on Sunday June 17, 2001 @03:59AM (#145732) Homepage
    Now when the aliens recieve our DNA [slashdot.org] they won't know if we have sent them our genetic blueprint or just our latest symphonic masterpiece.

  • That's just great.. (Score:3)

    by gabriel_aristos ( 265988 ) on Sunday June 17, 2001 @04:02AM (#145733)
    So now, not only will companies be able to patent specific genes, they'll be able to copyright them too? That's all we need.. the RIAA to get involved in our DNA.

  • umm.. who says this will sound good or even remotely like what we call music...
    Pretty cool though, considering they try to map Mozart's music with fractal geometry and many believe that DNA could use fractal geometry instead of having to store all the neccessary info... Maybe a few of our DNA strands will turn out to be Chopin or Tchaichovsky... hehe

  • This subject tweaked my memory - it turns out that the father of one of my high school pals worked in this area as well, in the 1981 time frame. Susumu Ohno worked at the City of Hope Cancer Center. His obituary, which mentions this work, is at

    http://www.cityofhope.org/cohpress/Medical_Scien ce /Ohno_obit.htm

    A web search on "Susumu Ohno City of Hope Music" yields numerous other citations, including this one:

    Dr. Susumu Ohno, a geneticist at the Beckman Research Institute of the City of Hope in Duarte, California believes that the elements of nature have a unique series of organized identifying pitches. He assigned a musical pitch to the elements (do)Cytosine, (re mi) adenine, (fa sol)guanine (la ti) ihymine, that exist in creatures, thereby making melodies. All creatures have different melodies. When played before professional musician, they believed them to be melodies created by Bach, Brahms, Chopin and other masters because of their intricacies and organization. Even cancer cells have their own "melodies" The Body of Music Chapter 13 page 141 "Meaning and Medicine" Larry Dossey, MD Bantam

    As I recall, he used a Macintosh to produce the music.

    ...Mark
  • metamath.org has something similar to this with math proofs if you don't want to put SlowJava on your pc. I'm sure Slashy-D mentioned this somewhere... I just don't care to search... morning time... can't see computer... falling into stream of consciouseness...and loose spelling
  • Say we take some DNA, and use it as a protein blueprint for some kind of organism, then pick the most pleasant sounds it produces and call that music?

    Or has that been done?
    --
  • DNA Evidence [howstuffworks.com] is really cool, and so are DNA Computers [howstuffworks.com].
  • RIAA, "Sir, we've discovered that your DNA is actually an N'Sync song, and as such we now own the rights to your body and reserve the right to collect royalties on any copies of your DNA you distribute, not withstanding skin cell loss, hair loss, and and time you ejaculate as this would constitute illegal copying of our music."
  • ..lf l'm ever stranded on Mars, trying to get into that big face, l can bring this thing along.

    (lts not a spoiler, since l didn't say what movie that was ;).
  • it also reminds me of another DNA... doug adams. In his "Dirk Gently's Detective Agency," one character is obsessed with his idea that all natural phenomena should somehow translate into "beautiful" music. interesting what it turns out in the end the author seems to think comes closest...
  • I recognize this as a huge scientific accessibility advance. Now blind people can easily access DNA. I wonder how good has the sound hardware to be for one be able to distinguish differente DNA.

    Now you can see DNA models, you can hear DNA. When will it come the time when you can finally feel information ?

  • Here are a few more (possibly absurd) considerations to make?

    What would happen if a section of the music derived from your DNA just happened to match a well known musical hit? Would you be getting nastygrams from the RIAA? (Another reason for certain groups to lobby for a repository for everybody's DNA, perhaps).

    In this unlikely case, who would own the copyright? The original artist / publisher, or me? What about prior works? Had I been born (or possibly even conceived) before the music was composed, would I have a legal leg to stand on? Or would I have to pay royalties, and at what point? Every time a cell divides, maybe. It could be split down to easy to make monthly payments, perhaps, determined by the average rate of cell division.

    And don't you go giving your blood away either. It would be interesting to see the RIAA going after blood banks with the same force as Napster. Or maybe I would just be banned from giving blood instead.

    Also, what if there was a program to convert DNA into code. Could the same copyright issues apply then? Maybe this explains the rift between Linux and Windows users?? Genetic predisposition maybe? In fact, if there is a method of producing some form of meaningful data from DNA, can you treat the two as the same?

    I am reminded of the prime number that when un-gzipped produced decss [slashdot.org], bringing up the issue of whether a number (or in this case, DNA) can be treated as the same as something that is decoded from it. Surely there are an almost infinite number of ways to decode DNA. If just one of those methods decodes a copyrighted work from DNA, what then?

    --please insert witty .sig here --

  • I can send predominately random notes into a MIDI player, based on DNA, just so I can produce about the same quality music as I could with a Random Number Generator.
    ----

  • Java (Score:1)

    by werbz ( 457972 )
    I have a prolblem with java being platform independant... It isn't... Since you need (a) java virtual machine to interpret it. Check me if I'm wrong (please) but if I whipped up my own platform here (let's say the werbzx87) I'd have to develope a java VM too!
  • Now we can have our own personal theme songs! If it happens to sound like trash, well, that's your fault.
  • I don't want to pick nits, but there's TONS of room for creative input in the world of 12-tone music. Schoenberg's matrices were only that - a collection of potential row forms that were related but could be freely deployed in a work.

    Later on, of course, the so-called "post-Webern" serialists tried to apply that sort of organizational scheme to additional parameters (orchestration, rhythm, timbre, etc.)

    But choice never went away. The closest we got (maybe) was in Xenakis's computer-generated music based on stochastic theory or in Cage's chance music.

  • The group shamen did this several years ago, on the DNA sequence coding for the 5HT-2a receptor.
  • Will people be able to be sued for playing their own DNA music? Companies that have already patended certain DNA sequences could sue you for divulging trade secrets! How will the courts deal with this? Who is the author? You? Them? God? I'm sure the corporations and the RIAA will have their lawyers hopping on this one. This would be an aweful predicatment. Everyone loses, but in the end how good can DNA sound?
  • If you would have read the web page, you might have noticed that the program was developed co-operation with original developers!

Slashdot Top Deals

"Everything should be made as simple as possible, but not simpler." -- Albert Einstein

Close