I've recently read a couple articles dealing with long term data storage, and DNA was suggested as one of the prime candidates for long term storage of digital data additionally, there is this article which deals with decoding of long storage data. These popular articles, though simple do mention important concepts- how packets are stored in DNA use addressing schemes, redundancy of storage, writing for a recipient who may be using a different language, etc.

This made me think - we have the SETI initiative which is looking for signs of (intelligent) life in the universe. That organization was created long time ago, and uses technologies of those days - radio waves. The modern science knows that we can store a lot of data in DNA, and that most organism DNA contains only a small fraction of protein coding sequences and a large amount of non-coding material. If I understand correctly, some of the non coding material is repetitive and highly preserved. Has anyone considered expanding the search for intelligent life to include examination of DNA for messages that might've been left by other intelligent species actually visiting earth?

  • 2
    $\begingroup$ You suggest an interesting idea, but what's the question? $\endgroup$
    – Drosophila
    Mar 27 '13 at 17:56
  • 1
    $\begingroup$ Unfortunately most of the highly conserved non coding material that you mention comes from viruses which are not really prime candidates for intelligence. $\endgroup$
    – terdon
    Mar 27 '13 at 18:17
  • $\begingroup$ The question is about bioinformatics - is anyone looking for "long storage" data within DNA sequences. $\endgroup$
    – Alex Stone
    Mar 28 '13 at 1:53
  • $\begingroup$ +1: I had the idea about DNA as database too, but it didn't occur to me to marry that thought with distributed computing initiatives! Well done! $\endgroup$
    – Everyone
    Mar 28 '13 at 18:46

Let's extend your idea a bit... Ok, there are conserved sequences that we may not know what their function is. Let's assume that they are indeed not functional and are some kind of message left by an ancient form of intelligence.

How would you go about detecting that? It is already given that they are highly non-random, but this is not surprising and many "non-message" sequences have this property. Also consider that these sequences are quite short, meaning that they will have low information content. This means that if you check enough conversion codes and use imagination, you would probably be able to find several "messages" which are "hidden" there (e.g. "bible codes" etc). In other words, as I am a scientist that deals with probability and pattern recognition frequently, you will have a very hard time trying to convince me that you found a real hidden message... Formally, you don't have a satisfying background model.

The main difference from SETI is that there they know that their background is essentially random noise, so it is much easier to detect "intelligent" messages.

  • $\begingroup$ Very good point. What originally inspired the question was that the "data in DNA" encoding linked in the question uses very short 96 bit "packets", preceded by a 19 bit addressing sequence. While the payload can vary, the presence of the address within each packet gives hope of detection. As far as I understand, this is how genes are detected - by their promoter/terminator sites. $\endgroup$
    – Alex Stone
    Mar 28 '13 at 3:42
  • 1
    $\begingroup$ DNA in living creatures is just not a very good storage method. Critters can go extinct or mutate your 'message'. Theoretically you could plant a sequence of non-coding DNA in a large number of completely unrelated organisms, and that consensus sequence would last a little while. It would decay pretty quickly without selective pressure, though. In fact, you could look for these right now. Go BLAST species that are unrelated looking for highly-conserved noncoding sequences. Worst case scenario you discover a new method of gene regulation or something. $\endgroup$
    – Resonating
    Jun 23 '14 at 14:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.