RE: HUMANS and low genetic complexity

From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Fri Jul 11 2003 - 07:11:48 MDT

  • Next message: Robin Hanson: "Re: Bostrom & Hanson in Article"

    On Fri, 11 Jul 2003, Jeff Davis wrote:

    > --- Spike <spike66@comcast.net> wrote:
    > > This irrascible immunologist clearly was
    > > never a programmer. Had he been a software
    > > developer, it would be clear to him how it
    > > happens that pieces of dead code can remain
    > > in the final product, even after many versions
    > > of the software. There would need to be some
    > > evolutionary pressure to reduce or darwinnow
    > > the dead code, which there evidently isn't.
    >
    > I'm with Simons on this one. Ever since the term
    > "junk dna" gained currency, I've wondered why someone
    > hasn't challenged the notion. [snip]

    I'm afraid I have to be a wet blanket on this one.
    At least 50%, and perhaps as much as 95%, of the human genome
    (probably most mammalian genomes) really is indeed "junk" [1,2].

    > Can it really be said--having barely embarked on the
    > journey of discovery-- that we know enough about the
    > function of dna to declare the apparent 'excess' to be junk.

    Much of it is "junk" from a currently useful genome standpoint.
    It is not however "junk" from an evolutionary (& selfish gene)
    standpoint. That is because much of it consists of interspersed
    repeats, especially LINEs, SINEs, LTRs, etc. (see pgs 879-888 of
    [2] for a very good explanation with nice statistics).

    > Nature 'believes' in efficiency. Fitness and
    > efficiency go hand in hand. It took an awful lot of
    > energy to make all that 'junk'.

    This is too simplistic Jeff. Efficiency can be measured in
    many ways. In mammalian genomes it seems to be measured
    by selfish genetic elements in a way something along the
    lines of "how fast can I copy genes to increase the pace
    of evolution". In Fugu (pufferfish) genomes it seems to
    be measured in a way along the lines of "how much junk
    can I remove from the genome without destroying functions
    critical to the survival of the species".

    That is why Fugu can produce a vertebrate with ~430Kbases
    of DNA while it requires most mammals almost an order
    of magnitude more to accomplish the same task.

    > The utility of the 'extra' dna awaits
    > the gradual elaboration of the grand scheme.

    Probably not. The only "utility" speculations I've seen
    this far may be in helping to align similar chromosomes during
    meiosis, mitosis and homologous recombination DNA repair.
    But this has a down-side -- the similarity between many
    of the transposable elements (interspersed repeats) allows
    these processes to become confused (thinking two regions
    of a chromosome pair are the "same" region when they really
    are not). This can produce insertions and deltions in
    chromosomes leading to diseases like DiGeorge and
    Williams-Beuren syndromes [2, pgs 911-912].

    > I've heard tell that for birds, evolution winnowed the
    > cargo of dna, the better for the birdies to get and
    > stay airborne. Or so the theory goes. If indeed
    > evolution trimmed their dna, then we have a proof of
    > principle re dna trimming for fitness purposes. The
    > profligate and wasteful production of 'junk' dna
    > similarly seems an idea too heavily freighted to get
    > off the ground.

    I once thought this as well (may have even said so on the
    list sometime) but Robert Freitas has argued fairly
    convincingly to me that the mass of the DNA is negligible
    relative to the mass of the cell itself so there is
    unlikely to be significant selective pressure to
    reduce the quantity of DNA in animals that fly.

    There doesn't seem to be much selective pressure at all
    on carrying around (and having to replicate) what is
    probably lots of 'junk' DNA -- if there were then the
    Amoeba dubia wouldn't have 700 pg of DNA (~200x the
    human amount for a genome size of ~670 billion bases) [4,5].

    > Some time ago, when I first got annoyed at the seeming
    > presumptuousness of the 'junk' dna concept, I wondered
    > if perhaps the 'junk' might be preferentially
    > distributed in the outer layers of the nuclear
    > package, providing a 'shield' for the really vital
    > 'inner' bits. You know. Whatever nasty dna-degrading
    > agencies--free radicals, mutagenic compounds, ionizing
    > radiation--runnin' around loose in the cytosol, would
    > have to get past the protective/sacrificial outer
    > layers before thay could do any 'real' damage. Just
    > an idea.

    Though it isn't distributed as you descibe, the concept
    does have some merit -- free radical damage, double strand
    breaks, etc. are likely to be much less harmful if they
    take place in the "junk" rather than in the "useful" DNA.
    Since most, if not all, of this type of damage impacts
    DNA randomly retaining random "junk" may have a slight
    benefit for personal and species genome stability. But
    this isn't a very strong argument -- mice and humans
    have a 30-50x difference in longevity and yet the amount
    of "junk" in their genomes is similar (mice have only a
    slightly smaller genome than the 3+ gigabase human genome).

    Robert

    1. Gee, H. "A journey into the genome: what's there",
       Nature Science Update (12 Feb 2001).
       http://www.nature.com/nsu/nsu_pf/010215/010215-3.html
    2. Int. Human Genome Sequencing Consortium, "Initial Sequencing
       and analysis of the human genome", Nature 409:860-921
       (15 Feb 2001).
       http://www.nature.com/cgi-taf/DynaPage.taf?file=/nature/journal/v409/n6822/abs/409860a0_fs.html
         (careful -- the article is big)
    3. A number of articles in Science Vol. 291 (16 Feb 2001).
       http://www.sciencemag.org/content/vol291/issue5507/index.shtml
       esp.
       http://www.sciencemag.org/cgi/content/abstract/291/5507/1304
       http://www.sciencemag.org/cgi/content/full/291/5507/1304
         (careful -- the article is big)
    4. Animal Genome Size Database: Summary Information
       http://www.genomesize.com/summary.htm
    5. Genome Size slide:
       http://www.brc.dcs.gla.ac.uk/~drg/courses/bioinformatics_mscIT/slides/slides2/tsld006.htm



    This archive was generated by hypermail 2.1.5 : Fri Jul 11 2003 - 07:22:56 MDT