[LMB] (Slightly) OT: Mitochondria (No spoilers) back on topic

Dan Tilque dtilque at comcast.net
Fri Aug 10 12:04:28 BST 2007


Tora K. Smulders-Srinivasan wrote:

>
> Though, I gotta say, mitochondrial inheritance is a very very
> unlikely manner of actually passing on genes.  There are
> exactly 13 polypeptides encoded by the mitochondrial genome
> ...  So it's highly unlikely
> that a gene for groundsense would be slopped into the
> mitochondrial genome -- it's a tiny piece of DNA (only ~1.6
> kb) and very highly selected and very stripped down and
> without lots of junk DNA as you get in the nuclear genome.

All, true, but there is another way to have inheritance only from
the mother or only the father. It's called genomic imprinting and
you can read all about it here:

http://en.wikipedia.org/wiki/Genomic_imprinting

Note the line: "No naturally occurring cases of parthenogenesis
exist in mammals because of imprinted genes." That explains why
some lizards and fish reproduce by parthenogenesis, but not, say,
shrews or rabbits. It also explains why virtually all cloned
animals have problems of some sort. Obviously, they are going to
have to figure out how to tinker with the imprinting to get the
clones to develop properly.

As far as the "junk" DNA goes, I (and I'm not alone) suspect that
a large fraction of it (about 98% of the DNA in human chromosomes
is not part of a gene) does have some function. Not all of it;
some of it is most definitely junk: large sections of repeated
patterns; pseudogenes; disabled viruses; etc.

My guess is that quite a bit of it helps regulate the production
of proteins. There are some 90K different proteins in the human
body produced by about 20K genes. (Each gene codes for about 4.5
proteins on average by means of various introns.) Chances are
most proteins have their own unique set of times, places, and
frequencies where they are produced. There's probably a large
number of mechanisms that control this (we've discovered a number
of them in the past few decades) and those in turn may be
controlled by "junk" DNA.

--
Dan Tilque





More information about the Lois-Bujold mailing list