Re: everything you know is wrong--again

Date: Sun Feb 03 2002 - 19:01:01 MST

In a message dated 2/3/02 3:28:00 PM Pacific Standard Time, writes:
>On Sun, 3 Feb 2002 wrote:

>To make it even harder -- answer the question --
>When you make an error copying DNA, such that you
>have base mispairings, how
>does the repair machinery know *which* strand is wrong?

The new strand isn't yet properly methylated.
Repair enyzmes can alter the unmethylated strand
before the methlation-copying enzymes get to it.
(Not to exclude other ID methods unknown to me or
even humanity in general, at present)

>> The real mystery to me is why efficiency
>>is seemingly not a factor
>> and why junk DNA is allowed to proliferate.

>Copying it is relatively cheap
>compared to the cost of deleting
>it and getting it wrong.

Right; DNA is only about 2% of dry weight.
The extra energetic costs are small, and to
fix it requires millions of advantageous mutations
(one to ditch each piece of junk.) So the per
fix advantage is very small - 1 x 10^-8 or so.

>transposons allow easy duplication
>of DNA that can subsequently evolved
>for extended uses retaining
>them in the genome allows faster
>evolution and probably selects
>for their retention.

Transposons select for their own retention.
There's millions of them in your genome,
probably hundred of active ones, all evolving
against control mechanisms, plus others dropping
in from time to time from other species or
from retroviruses. It's like asking why we're
not immune to viruses, really.

>Better questions are whether transposon
>activation is keyed to
>stress as the SOS response is in bacteria

Maybe. Theory suggest they should only replicate
in germ-cell lines, which is rather different from
stress. It's a plausible hypothesis of mine that
increased sexual activity will rev up accidental
non-germline transposon activity and this is a model
for the inverse relationship between sexual activity
and lifespan and possibly the CR effect.

>and how pufferfish (Fugu)
>have managed to keep the junk DNA
>so low (they only have a 400 MBP
>genome which is very small for a vertebrate)?

Yes, that's a doozy! And they did it by making
each stretch of junk smaller, but retaining (roughly)
the # of introns per gene. There's a guy here in
UC Riverside working on that a little - but he says
it's very hard because aligning the introns with
relatives is very ambiguous.

This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 13:37:37 MST