D
David Brown
Guest
On 03/02/2022 23:39, Joe Gwinn wrote:
Yes, indeed - but it is rare that they end up actually passing on
successfully to future generations, and rarer still that this leads to
useful new characteristics. It\'s a slow game!
I hadn\'t heard of that one - thanks for the link. This is similar to
how organelles like mitochondria might have evolved from inter-cellular
bacteria.
As I said. It happens, but very rarely.
I\'d say that\'s a somewhat naïve and unhelpful viewpoint. I\'m sure you
have heard the term \"survivors\' bias\" ? That\'s what happens in
evolution. Evolution does not \"solve\" problems. It causes gradual
changes (with /very/ occasional jumps), with a selection towards species
with more long-term success. We can look species alive now and claim
\"they solved the problem\", but really they are just the ones that are
lucky enough to have survived.
We can take a very over-simplified (it\'s missing any form of inheritance
to change long-term biases) analogy, and suppose you have a large
handful of dice and want to see which ones are good at solving the
problem of getting high scores. Roll them all, then throw out all those
that failed to get a five or a six. Take these and roll them again, and
throw out the ones to fours that \"died\". You are left with dice that
solved the problem and scored at least 5 consistently. In fact, the
majority of them will have had at least one 6. And note that none of
the dice was \"intelligent\", or knew the target characteristic or the
\"problem\" they were trying to \"solve\".
I know /you/ understand this, but Larkin does not appear to. He thinks
evolution has targets and aims, and that it picks the best way to \"move
forward\". If you have other ways to help him understand, then that
would be great.
It is an /analogy/. It is not a /model/.
On Wed, 2 Feb 2022 20:21:18 +0100, David Brown
david.brown@hesbynett.no> wrote:
On 02/02/2022 17:51, jlarkin@highlandsniptechnology.com wrote:
On Wed, 2 Feb 2022 16:29:53 +0100, David Brown
david.brown@hesbynett.no> wrote:
On 02/02/2022 14:49, jlarkin@highlandsniptechnology.com wrote:
On Wed, 2 Feb 2022 12:11:03 +0000, Tom Gardner
spamjunk@blueyonder.co.uk> wrote:
Viruses are an important vector for evolution. They are a common source
of horizontal gene transfer in microbes. Mutation or gene transfer via
viruses does not happen often in higher organisms
Are you sure of that? Horizontal transfer is a huge boost to
evolution, and everything has to evolve. Dispersion of improvements by
family descent is very inefficient.
Retroviruses do in fact insert their genes into the somatic genomes of
their hosts. Some of it makes it into the germline genome over time.
Yes, indeed - but it is rare that they end up actually passing on
successfully to future generations, and rarer still that this leads to
useful new characteristics. It\'s a slow game!
We know this because there are lots of (usually nonfunctional) viral
genomes in the genome of all animals large enough to see. HIV is the
current poster child in humans, SIV in simians.
Some bacterial can also fiddle with host DNA.
.<https://en.wikipedia.org/wiki/Wolbachia
I hadn\'t heard of that one - thanks for the link. This is similar to
how organelles like mitochondria might have evolved from inter-cellular
bacteria.
No, everything does not have to evolve - there are lots of organisms
that have changed very little for perhaps hundreds of millions of years.
Horizontal transfers /can/ introduce new genes, but the /vast/ majority
of horizontal transfers are detrimental to the organism if they do
anything noticeable at all.
True, but relevance unclear. And in Biology, there is always an
asterisk or two.
You may have heard of the \"immune system\". One of its jobs is to
minimise virus infections - identifying and destroying virus particles,
and destroying any cells that get infected. The better an organism\'s
immune system (and even bacteria have immune systems to counter
viruses), the lower the opportunity for virus infections of any sort,
never mind ones that transfer genes.
Patronizing. Not effective.
To get a horizontal transfer via a virus, you need multiple things to
happen. First, the virus must infect one host\'s cell but make a
monumental screw-up during the infection, so that a bit of the host\'s
DNA gets caught up along with the virus DNA when the cell replicates the
virus. That happens, but it\'s extremely rare - and usually such
mistakes means no successful virus production. Then this modified virus
needs to get into another host, and have another monumental screw-up on
the part of the virus /and/ on the part of the host cell, where the DNA
fragment gets mixed with the host\'s DNA. And then the host cell must
eliminate the virus (so that it doesn\'t die by virus reproduction), and
the DNA change must be non-fatal to the cell. If this happens in the
target host\'s germ cells then the mutation will pass on to the next
generation (probably killing them long before they can produce
descendants of their own). Otherwise you\'ve made either a one-off cell
change that ends with the cell\'s death, or a tumour.
Horizontal transfer from viruses to big animals like humans does in
fact happen, as discussed above.
As I said. It happens, but very rarely.
And even if horizontal transfers /were/ as potentially useful for
evolution in the long term as you imagine, that does not mean it
happens. You have this bizarre belief that just because something is
useful or efficient (in your eyes at least), evolution will cause it to
evolve.
Even though evolution is blind, it does seem to solve whatever problem
is presented, because that\'s where the most stress is.
I\'d say that\'s a somewhat naïve and unhelpful viewpoint. I\'m sure you
have heard the term \"survivors\' bias\" ? That\'s what happens in
evolution. Evolution does not \"solve\" problems. It causes gradual
changes (with /very/ occasional jumps), with a selection towards species
with more long-term success. We can look species alive now and claim
\"they solved the problem\", but really they are just the ones that are
lucky enough to have survived.
We can take a very over-simplified (it\'s missing any form of inheritance
to change long-term biases) analogy, and suppose you have a large
handful of dice and want to see which ones are good at solving the
problem of getting high scores. Roll them all, then throw out all those
that failed to get a five or a six. Take these and roll them again, and
throw out the ones to fours that \"died\". You are left with dice that
solved the problem and scored at least 5 consistently. In fact, the
majority of them will have had at least one 6. And note that none of
the dice was \"intelligent\", or knew the target characteristic or the
\"problem\" they were trying to \"solve\".
I know /you/ understand this, but Larkin does not appear to. He thinks
evolution has targets and aims, and that it picks the best way to \"move
forward\". If you have other ways to help him understand, then that
would be great.
Evolution does not have aims, guides, or targets. It is not intelligent
or guided. It does not optimise, or reach ideal solutions. It does not
make huge leaps to new methods just because you think these might be a
good idea. It does not always eliminate bad traits and enhance good
ones. It does not necessarily lead to the best choices or the \"fittest\"
results.
Actually, evolution does in fact optimize. Where needed. See above.
(as you say, they have
to hit egg or sperm cells - and the great majority of mutations are
either lethal or have little significant effect). But they are vital
for getting completely new features into the population. The placenta
in mammals, for example, has been traced to virus gene transfer.
As for kissing - how exactly do you think that would transfer genes?
For one, by transferring viruses. Maybe even cells.
Do you not realise how ignorant that sounds?
Ad hominem. Not effective.
Let\'s take an analogy that might make it simpler for you. A cell is
like an electronics board, and its genetic code is like the schematic
and pcb design for the board. Do you think you can change the designs
of the boards in some machine just by putting a different board beside
them? That\'s your \"cells transferred by kissing\" idea.
Perhaps you should bang the different boards together and see what
happens - that\'s the horizontal transfer. One time out of a billion you
might make a short-circuit, or break a track, that gives one of the
boards new characteristics that you hadn\'t seen before.
Compare that to taking two extremely similar designs (99% or more
match), and swapping a few corresponding sections of the schematic to
see if there is an improvement. If you\'ve swapped something critical,
it will probably not work at all. If you\'ve swapped some values of
filter components, maybe you\'ll get a slightly better filter. That is
the analogue of sexual reproduction.
Which method do /you/ think is going to be more successful?
This is a straw man argument, and deeply flawed to boot. DNA and RNA
are recipe strings, and not schematics, so the analogy fails.
It is an /analogy/. It is not a /model/.
One can use Genetic Programming to evolve circuit designs. Was
interesting, but did not turn out to be all that useful in practice.
.<https://en.wikipedia.org/wiki/Genetic_programming
Evolutionary Design of Digital Circuits Using Genetic Programming
.<https://arxiv.org/abs/1304.2467
If you don\'t want that random brain-fart to be dismissed, you\'ll have to
justify why you think there is significant genetic material in spit,
Dismissed? Ideas should be played with. But I don\'t mind if you
instinctively attack ideas... less competition.
I offered you a chance to justify your idea before I dismissed it. It
turns out you had nothing. Good ideas are useful - but your view that
all ideas are somehow worthy for consideration is absurd. (I didn\'t
dismiss it out of hand - I thought about it, then dismissed it.)
People are put in prison based on the genes in a bit of spit.
Yes - so what? Do you think the DNA in someone else\'s spit magically
transforms you? Seriously? Do you also think that if you eat chicken,
you might sprout feathers? (Are you going to dismiss that idea out of
hand?) If a radioactive spider bites you and injects some of its DNA in
its saliva, is that going to turn you into a superhero? (Surely you
will play around with that idea too, to give you an edge on your
competition.)
Males kiss, as kissing the hand or ring of a king or the pope. Some
cultures kiss a lot more than we do.
And some do so less. None of that matters.
If kissing spreads disease, it would be deselected. But it\'s not. Same
with shaking hands... spreads germs!
Kissing /does/ spread disease - as does shaking hands, and any kind of
contact. But the benefits usually outweigh the risks.
Well, to be precise, what is deselected are people who cannot handle
the close contact needed to be a species of social animal. Immune
systems also evolve to suit.
You are talking past one another. Your objectives are different, so
both can be correct, or not, independently of one another.
Joe Gwinn