Re: In the news: Researchers Predict Infinite Genomes


John Harshman wrote:
> Glenn wrote:
>
> > <rja.carnegie@xxxxxxxxxx> wrote in message
> > news:1127613914.409390.187040@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
> >
> >>Glenn wrote:
> >>
> >>>http://www.sciencedaily.com/releases/2005/09/050923075708.htm
> >>>"TIGR scientists have concluded that researchers might never fully describe
> >
> > some
> >
> >>>bacteria and viruses--because their genomes are infinite. Sequence one strain
> >
> > of
> >
> >>>the species, and scientists will find significant new genes. Sequence another
> >>>strain, and they will find more. And so on, infinitely."
> >>>"Many scientists study multiple strains of an organism," says TIGR President
> >>>Claire Fraser. "But at TIGR, we're now going a step further, to actually
> >
> > quantify
> >
> >>>how many genes are associated with a given species. How many genomes do you
> >
> > need
> >
> >>>to fully describe a bacterial species?"
> >>
> >>Once all possible combinations of DNA codes are exhausted,
> >
> >
> > How many are there?
>
> Hard to say. In theory, there are infinitely many possibilities. But
> this requires that a string of DNA can exist that's infinitely long.
> Obviously, that's absurd, but neither can we easily say what the longest
> possible DNA string is. So the boundary of possible lengths is nebulous.
> But if we have the longest possible string figured out as N, then the
> number of distinct genomes of that length is 4^N. Shorter genomes are
> all subsets of those, but if we count them as distinct then we have to
> add 4^(N-1), 4^(N-2), etc. There's a simple mathematical description of
> the resulting number, which I leave as an exercise for the reader.
>
> >>the Designer
> >>will no longer consider, "What creature shall I create today?" But
> >>"Which creature shall I create today?"
> >>
> >>After Lewis Carroll, _Sylvie and Bruno Concluded_.
> >>
> >>"But lunatics would always write new books, surely?" she went on. "They
> >>couldn't write the same books over again!"
> >>
> >>"True," said Arthur. "But their books would come to an end, also. The
> >>number of lunatic books is as finite as the number of lunatics."
> >>
> >>On reflection, Carroll's Lunatic Book Conjecture actually runs into a
> >>Turing Machine Halting Problem, but I don't think that applies to
> >>bacteria or viruses, which have an upper size limit.
> >>
> >
> > What's the limit?
>
> It's not a logical limit but a physical limit, and there's no simple way
> to decide it. What limits are we placing on bacterial metabolism? Could
> there be a bacterium as big as a pea, with room for a very big genome
> indeed? I doubt it, but who knows what tricks of metabolism might make
> it possible? Would the largest known bacterial genome do? The biggest
> one I can find is a Nostoc genome at 10 megabases.

Should I apologise for bringing Lewis Carroll into this? Seriously.

The mathematics is genuine, but not realistic. If I'm doing this
right, then taking 10 megabases as - for the sake of argument - the
largest possible size of DNA, then it would take 2.5 megabytes just to
STORE the NUMBER of different possible combinations of DNA. It is a
finite number but it is VERY VERY VERY large. For comparison, the
number of different arrangements of a pack of 52 playing cards can be
stored in 8 bytes (if I did /that/ right).

One usually makes, at this point, a comparison unfavourable to the
number of grains of sand in the Universe.

Of course many of these vastly numerous hypothetical genomes are not
significantly different to others - certainly not, given what was
already discussed, different species. And, of course, almost all of
these /random/ "genomes" are not fit for living organisms at all.

But still, the number is very vast; and although it is finite, it is
probably too vast to be exhausted, and so for /practical/ purposes is
almost infinite.

This would be the case even without horizontal transfer, and /with/ it,
perhaps there is an argument for a very little less emphasis on
studying bacterial species, diverting attention to studying individual
bacterial genes, whichever particular organism they may happen to
reside in?

.

Relevant Pages

  • Re: In the news: Researchers Predict Infinite Genomes
    ... > this requires that a string of DNA can exist that's infinitely long. ... In one genome. ... >>>bacteria or viruses, which have an upper size limit. ...
    (talk.origins)
  • Re: How to account for life?
    ... I know little about biology, ... Nature tried randomly to obtain that long 3 billions string every ... the odds against our human DNA to be formed by chance ... There is no way that the DNA (even of the simplest bacteria) ...
    (talk.origins)
  • Re: How to account for life?
    ... I know little about biology, ... Nature tried randomly to obtain that long 3 billions string every ... the odds against our human DNA to be formed by chance ... There is no way that the DNA (even of the simplest bacteria) ...
    (talk.origins)
  • Re: How to account for life?
    ... I know little about biology, ... Nature tried randomly to obtain that long 3 billions string every ... the odds against our human DNA to be formed by chance ... There is no way that the DNA (even of the simplest bacteria) ...
    (talk.origins)
  • Re: Copes rule and bacterial evolution
    ... These early forms of life were not very good at filling eco-niches ... that some DNA leaking from one would get into the other, ... selection pressure might merge genomes, or have one drive the other to ... RNA was too fragile), and how DNA was retrieved to restore RNA later ...
    (sci.bio.evolution)