|
A genome packaged as "inscribed matter" is discussed
only briefly in the paper linked below.
"There have been some speculations that a simple
biological system carrying a message and capable of self-replication in suitable
environments may be one possible channel for interstellar communication (Yokoo
and Oshima, 1979, and Nakamura, 1986). These kinds of ideas have several and
severe objections. For example, the impossibility of predicting the environment
of the target star in order to favor the self-replication of the molecular
structure, the impossibility of avoiding the destruction of the content of the
message by molecular mutations, and the impossibility for us to distinguish
between a "natural" organism and a real biological interstellar
message."
If a "Common Europoid" hypothesis can be supported
(and I think I have at least the outlines of a case), the "impossibility of
predicting the environment" objection may not have quite as much force.
It'll still be impossible, of course, but hey, it's impossible to cross a busy
street on a "walk" signal without risk of getting run over, but we still do it
all the time because it works almost all the time.
The recent discovery of stable "junk DNA" would seem
to knock down the mutation objection, unless by "molecular mutations" the author
is referring to cosmic ray interference - a problem that interstellar shielding
could solve -- or quantum positional uncertainties over long time scales causing
the inscribed signal to 'rot', which I think is either ridiculously improbable,
or solvable with an active transport that regenerates the message using
error correcting codes.
The same "junk DNA" discovery - that you can erase
all of the silent DNA from of a mouse sequence, and you still leave a recipe for
a normal-seeming mouse - suggests that the problem of distinguishing between
life evolved in the solar system and life prepared outside the system may not be
so hard. It reduces to something like the garden-variety SETI problem,
once you've identified a candidate: how do you tell if that "stable junk DNA" is
just noise, or something else, and if it's something else, is it intelligently
fabricated? I assume that, by now, someone has looked at genomic
information currently on hand for obvious markers - like pi out to 200 bits or
something like that. (Even just "zero out to 200 bits" would be enough,
right? It just has to be obvious and highly improbable.) So maybe
genomic SETI is easily declared a dead-end. But maybe it's subtler than
that, and subtle for some reason we haven't figured out yet. Maybe
the very fact that there ARE stable junk sequences is a hint to us that they
aren't junk.
It could be that I've been writing memory-efficient
code for too many years of my programming career, but it seems to me that
evolution would tend to optimize out useless bits, not evolve to preserve
them. "Intelligent Design" dogma may be the least of our problems - "Blind
Watchmaker" is fine with me, until someone proves otherwise, but what are
we to make of "Blind Rube Goldberg", i.e., of a nonsentient process that
seems a little crazy and stupid, like a programmer who pads out his programs not
just with commented-out code (quite understandable, if overdone in practice),
but also with inline comments full of word salad? I don't know enough
about genetics to assert confidentlly that this is what the discovery of stable
'junk' DNA amounts to. I wish I did know. The only explanation I can
think of is a hazy memory of a paper by a guy writing about energy and
computation, in which he said that we should keep in mind that erasing bits
costs energy too. Collecting garbage isn't easy either. It may be
that efficient memory management is beside the point in evolution, given the
vast amount of space to work with.
-michael turner
----- Original Message -----
Sent: Saturday, September 04, 2004
5:01 AM
Subject: Re: ET bits in "junk" DNA
(was Re: A response to Rose and White's paper)
I think you will find this article interesting. Scroll down to near
the end before the bibliography:
Larry
----- Original Message -----
Sent: Friday, September 03, 2004 9:01
AM
Subject: ET bits in "junk" DNA (was Re:
A response to Rose and White's paper)
> Michael:
>
> What do you think of Paul
Davies' hypothesis in "Do we have to spell it
out"
> (New Scientist
vol 183 issue 2459 - 07 August 2004, page 30 -- see excerpt
> below)
that ET might have inserted a message into highly
conserved
sequences
> of junk DNA? Note that further support for
panspermia is provided in
"Alien
> microbes could survive
crash-landing" (Nature online: 02 September 2004;
>
doi:10.1038/news040830-10), although some form of cosmic ray/gamma
ray
> protection would be required for extrasolar-system
propagation.
I think Jack Reeve mentioned something about this
earlier on the list.
I love this idea more than life itself (um, as
it were.) When people ask
me if I think aliens ever have visited,
or will visit, I say, "Either never,
or long before we evolved."
Aliens could be all ready to bloom
(or metastasize, if you're the
paranoid type) as soon as we decode them
from the global genome. Or
there may be just some interesting message
in the DNA bottle.
>
Davies said:
>
> So might ET have inserted a message into the
genomes of terrestrial
> organisms, perhaps by delivering carefully
crafted viruses in tiny space
> probes to infect host cells with
message-laden DNA? It's an idea that has
> been swirling around for a
few years, and has recently been championed by
> the Apollo astronaut
Rusty Schweickart. But on the face of it, there is a
> serious
problem. Living cells are not completely immune to
change.
Mutations
> introduce random errors into the stored
information, and over a long
enough
> time span they would
inexorably transform ET's message into molecular
>
gobbledygook.
True enough I guess, and with a long enough transit
time, maybe there's even
the danger of mutation from simple Heisenberg
position uncertainty, even if
bugs
were adequately shielded from
radiation somehow. Though I wouldn't
know where to begin computing
those probabilities. Maybe they are
negligible after
all.
> To minimise the effects of mutations, it would make sense
to incorporate
the
> message into a highly conserved segment of
DNA. Such segments are normally
> associated with key coding regions
of the genome that control the most
vital
> functions of the
organism. They tend to be unchanged between species,
> suggesting an
ancient origin. Mutations in such regions are invariably
> fatal. But
unfortunately tinkering with them by inserting alien DNA would
>
likely prove as lethal as any random mutation.
Yes, from *mutations*.
But there may be more than one way to rearrange
features
stably across
generations to the same effect - an arrangement might convey
bits
somehow, if there are enough degrees of freedom. Think of a
computer
program that can be written a number of ways. A mutation
is like
changing one character randomly, and if that's not changing
meaningless
whitespace to other meaningless whitespace, the program's
gonna crash,
or do weird things (or not even compile). However, if
you swap two lines
that have no execution-order dependencies, you're
OK. Could you
steganographically encode a message in a program by
drawing from
the space of possible behavior-preserving transforms of the
source code?
I'm sure you can. Can you do this with DNA as well?
I'm not so sure
about that. But ... read on ... it might be
irrelevant.
> Conversely "junk" DNA - sections of the genome that
seem to serve no
useful
> purpose - can be loaded with all manner
of genetic oddments without
> affecting the performance of the cells.
Inserting a message here would
> almost certainly be harmless. The
trouble is, junk DNA is famous for
> accumulating lots of mutations.
So the choice seems to be between killing
> the messenger and
compromising the message. What is needed is a region of
> junk DNA
that is also highly conserved.
>
> Until recently, this would
have been regarded as an oxymoron. But no more.
> Genomics researchers
at the Lawrence Berkeley National Laboratory in
> California who
compared human and mouse DNA have reported the discovery of
> vast,
highly conserved sequences of junk DNA (New Scientist, 5 June,
p
18).
> These segments are apparently surplus to requirements.
When the
researchers
> deleted them from the mouse DNA, the animals
seemed to be perfectly
normal.
> If ET has put a message into
terrestrial organisms, this is surely where
to
> look.
And
think of the money saved on giant radiotelescope arrays.
In any case,
for natural panspermia, I think we're talking about a
scenario in which
some object crosses interstellar space relatively
slowly, with adequate
shielding, maybe aerobrakes and gravity-
brakes across a lucky sequence
of gas-giants to reduce the
kinetic energy gathered from plunging toward
the sun, and
finally makes it through the atmosphere to Earth. (Or
Europa!
Back on topic! Finally!) That last part doesn't
strike me as unlikely
at all. The right kind of rock can plop down
in an ocean without
even getting very warm inside from passage through
the
atmosphere. That part's proven. The earlier parts of the
journey
are what bother me - I think to make panspermia LIKELY,
not
just POSSIBLE, you've got to have intelligent engineering
behind
it. But if someone can put reasonable numbers
in front of me
suggesting otherwise, I'm all ears. Or eyes.
Speaking of "vast,
highly conserved sequences of junk,"
I haven't cleaned my room for over a
month. So this should be
my last posting on this thread for, oh, at
least 24 hours.
OK, make that 12 hours. Less than a million years,
anyway.
-michael turner
[EMAIL PROTECTED]
==
You are
subscribed to the Europa Icepick mailing list: [EMAIL PROTECTED]
Project information and
list (un)subscribe info: http://klx.com/europa/
|
