TheWill
Originally posted by TheWill
You are still assuming that we HAD to have an insulin gene in the first place - we didn't.
My apologies if I gave the unintentional impression that "we HAD to have an insulin gene".
I "picked on" insulin purely because it was a convenient gene that with a length of 153 nucleotides, was not too small and not too big and would
readily lend itself to probability analysis.
Originally posted by TheWill
You are making the assumption that the insulin gene arose from non-coding DNA. It is just as possible that it arose from coding DNA, a series of bases
already coding for a protein which, by a few substitutions or perhaps just a single deletion/insertion, or even just a fortuitously placed crossover
event, would have changed the amino acid sequence produced by the gene.
If we place an "artificial" limitation on nature and stating that there is only one correct sequence of 153 nucleotides that would result in
insulin, then yes, the odds against are a whopping 10^92 (approx).
But as you have correctly pointed out, insulin is the end product of a degenerate coding system. So what happens to the odds now ? Do they plummet
dramatically and thereby allowing nature to easily evolve insulin ? Lets take a look.
You've calculated, based on degeneration, that there are (approximately) 10^24 ways of stringing 153 nucleotides sequentially and produce insulin.
Ok, fine .... adjusting the original odds of 10^92 with your 10^24 degenerate alternatives gives us new odds of
10^68 AGAINST
Thats a 1 followed by 68 zero's !!
But the end result is that we've all we've achieved is to simply exchange or downgrade incredibly stupendous astronomical odds against ... to simply
mind boggling astronomical odds against
Look, it's all fine to talk about high level activities such as degenerate coding, cross overs, substitutions, etc, etc ... but at the very basic
level of evolution lies nothing more sophisticated than nature's equivalent of a crap shoot.
It's
all about probabilities, chances and odds ... nothing more and nothing less.
What's the probability of any nucleotide being added or inserted ?
What's the probability of the added/inserted nucleotide being a specific one ?
What's the probability that the insertion is beneficial ? ... neutral ? ... harmful ? ... lethal ?
What's the probability of a nucleotide being added/inserted into a particular location as opposed to any other location ?
What's the probability of many nucleotides being added/inserted at a given location/time ?
What's the probability that a nucleotide sequence is created that codes for a viable protein ?
What's the probability that a number of seperate proteins can co-ordinate with each other to produce higher level activities e.g. Citric Acid cycle
?
Thain Esh Kelch
Originally posted by Thain Esh Kelch
You clearly don't get how things work. You seem to think that mutations are a random insertion of a nucleotide somewhere?
Errr, basically yes, thats exactly how "things work" ... coupled with random changes to existing nucleotides.
Perhaps you'd care to educate me on the
other ways that genomes increase in size over time if
NOT by strictly random means.
So according to you ...
You seem to think that mutations are a random insertion ...
if insertions (mutations) are
NOT random, then the logical implication must follow that there are times when insertions (mutations) are
deliberate - in other words, there is deliberate design and planning attributable to at least some mutations ?
Really ???? I think not.
Originally posted by Thain Esh Kelch
Without taking into account how many different ways mutations can be achieved, you should know that it is possible for entire chromosomes to be
duplicated - Double the genome in the life span of a single celled organism, which can be down to a few minutes. So your way of looking at time is
waaaaaaaaaaay off.
And just how complex can the genome of a "single celled" organism become and the organism still remain single celled relying on just mitosis ?
Mitosis is a type of cellular reproduction where a cell will produce an identical replica of itself with the same number and patterns of
genes and chromosomes.
So according to the above quote, within a single-celled organism, mitosis is a very efficient and effective way of maintaining genetic
stability.
In other words, your example of massive genetic growth in a single celled organism is therefore highly unlikely (there's that nasty probability
sneaking in again) and consequently, significant genetic drift is
minimal over very long periods of time.
Originally posted by Thain Esh Kelch
During mitosis, it is estimated that a mutation occours 1 in 10.000.000 nucleotides with efficient DNA polymerases, while we are down to 1 to
1.000.000.000 in higher mammals, that have proof reading polymerases. That is *roughly* 3 mutations per germ cell in higher mammals, which is quite
low.
As you've just admitted ...
That is *roughly* 3 mutations per germ cell in higher mammals, which is quite low.
the mutation rate decreases significantly over time as the organism evolves.
Which is my point exactly ... the observed average accumulation rate of 3.1 billion mutations over 3.8 billion years is inconsistent with expected
lowered mutation rates as the organism evolves. Taking your observation into account, we would have to conclude that for some unknown reason early in
evolutionary history, there must have been an incredibly accelerated rate of the initial mutation rate in order to compensate for the much later
falling of or reduction in mutation rates in higher organisms.
If you disagree, then what in your opinion would the "average" yearly mutation rate have to be to result in the final human genome consisting of 3.1
billion nucleotides if you have at most a maximum of 3.8 billion years in which to "generate" those 3.1 billion mutations ?
Originally posted by Thain Esh Kelch
The citric acid cycle ... is a series of enzyme-catalysed chemical reactions, which is of central importance in all living cells that use oxygen as
part of cellular respiration.
In eucaryotes yes.
Irrelevant ... as long as the organism employs oxygen for cellular respiration, then it uses some version of the Citric Acid cycle.
Originally posted by Thain Esh Kelch
The citric cycle is by no means perfect. It actually uses energy under some conditions, so again your argument is pointless.
Sorry but again your response is irrelevant.
Whether it's perfect or not, the Citric Acid cycle
does exist and is very complex and incredibly dependent on it's constituent sub-processes
working together as
one system.
Therefore my original argument regarding the odds for/against it's evolution is perfectly valid.
By the way, I notice that you have
not proposed a viable explanation for just how the independent proteins involved in the CA cycle came into
being ... let alone how these many proteins found themselves co-opted into working so intimately together.
And the odds of this (protein creation + working together = CA cycle) happening purely randomly is ...... (feel free to fill in a number here).
Originally posted by Thain Esh Kelch
This also tells that you don't *get* evolution. Even if insulin magically sprang into existence, it would not have an effect. You would also need
some way to regulate its expression, you would need a receptor it could target, and it would also need downstream effects from this receptor.
Thanks for adding additional "substance" validating my position.
Whether insulin is useful or not, whether it has an effect or not, is completely beyond the point.
We are discussing the nothing more than the possibility of the insulin nucleotide sequence coming into existence through purely random, probabilistic
means i.e. mutations. As shown many times, the odds against insulin arising through hit and miss, trial and error means is nothing less than
astronomical in value.
Now you're adding fuel to the fire by saying that not only did insulin evolve through random processes but now we have to also factor in the
probabilities that additional structures such as regulators and receptors (all protein products) would also have to spring into existence through
essentially random methods as well ... odds compounded upon odds ... compounded upon odds ...
rhinoceros
Originally posted by rhinoceros
It's so annoying when amateurs with no understanding whatsoever of genetics or microbiology in general make these ridiculous claims like the one you
presented.
Just to show how little you know. You said:
And like the insulin process, the Citric Acid cycle MUST have come into existence very early in the evolutionary time scale as it is critical to, and
used by all organisms that metabolize oxygen for respiration ... therefore giving nature very little time for trial and error.
This is just plain wrong. Ogygenic (oxygen generating) photosynthesis came about only about 2 billion years after life had started on our planet.
Before that there was no oxygen that could act as terminal acceptor of electrons, and thus no citric acid cycle as it's today.
I'll start my response with "I take offense to your condescending attitude".
So where was I wrong ?
The CA cycle as far as I know is used by every organism that uses oxygen for cellular respiration.
... which is of central importance in all living cells that use oxygen as part of cellular respiration.
Source: Citric Acid cycle
Therefore ... logic implies that the CA cycle must predate, or have evolved no later than the 1st organism that evolved the capability of using oxygen
for cellular respiration.
Anyway, here are the 3 possible scenarios ... by all means provide us with the wealth of your experience and knowledge.
Did the CA cycle evolve 1st and sit there waiting to be taken advantage of when the 1st oxygen metabolizing organism finally evolved ?
Or did the CA cycle evolve but only
after the 1st organism capable of metabolizing oxygen had previously already evolved ?
Or did the CA cycle
AND the 1st organism capable of oxygen metabolism evolve at the very same time in evolutionary history ?
Got to be one of the 3 above ... which one ?
Originally posted by rhinoceros
Originally posted by tauristercus
Ok, I can just barely, barely, barely accept that nature may just have some how fluked it with insulin ... but how do you explain nature doing it
another 25,000 times ... and some of those proteins are longer than insulin which means the odds against the longer ones is beyond imagination.
If you would actually read the posts in this thread you would know that this has already been explained.
Paper on the Evolution of Insulin.
Ok, just finished reading the paper and unless I completely misunderstood the basis of the paper, there was
NO mention whatsoever of an
evolutionary explanation of the
original and primary insulin type precursor gene.
Instead, the paper focused exclusively on the differences to be found within the insulin gene but only from the point of view of modern non-mammalian
vertebrates. In other words, how the insulin gene has evolved between species but not
HOW the insulin gene came into existence originally or
WHAT mechanism was responsible for the original insulin gene creation.
MrXYZ
Originally posted by MrXYZ
No it doesn't.
Look, I had game theory at uni as part of my MSc in Real Estate. If there's one thing I learned, it's that if something can happen, it eventually
will happen. To give you an example, winning the lottery is highly unlikely. In fact, depending on the type of lottery, you might only be likely to
win once in 256 MILLION years if you play once a week. Other lotteries are "easier" and you're likely to win once out of 120 million tries.
However, now look at how many people win the lottery every single week. It should be an incredibly rare event, yet it still happens.
Your chances of getting hit by lightning are approximately 2.5mil to 1, which is still a lot higher than you winning the lottery...yet people say
getting hit by lightning is super rare.
Unfortunately, in the above, you're using examples based on very small probabilistic ranges.
Lottery odds and your lightning example are both typically in the 10^6 possibility range.
In other words, the chances of selecting the correct sequence of numbers or being struck by lightning both have odds of just a few million to one
against the player (or victim).
Again, take insulin which is a relatively simple sequence of 153 nucleotides.
The odds of randomly building the correct 153 sequence has odds against it in the range of 10^68.
Comparing your lottery odds of 10^6 against the insulin gene odds of 10^68 is like comparing a drop of water to an ocean.
Originally posted by MrXYZ
reply to post by tauristercus
Please enlighten us how you came up with that figure. Why at the power of 90? Why 8x10?
I'm actually pleased that you asked the above question as it's pointed out a very minor error in my original calculation (big fingers on the tiny
calculator keyboard !). But having said that, this minor error in no way invalidates my original premise and in fact, continues to support it.
The calculation is easily done.
Lets start by using a simple variation of the classic 6 sided die but our version has only 4 sides representing the 4 possible nucleotides ... labeled
A, C, G and T.
If we roll the die once, we have a 1 in 4 (0.25) chance of guessing the outcome. No big deal.
Now, lets say we will roll the die twice in succession and we are looking for an A followed by a C as the outcome.
To calculate the probability of actually getting the A and C is simply a matter of multiplying 0.25 x 0.25 which gives an overall probability of
0.0625 or 1 chance in 16 of getting it right.
Similarly, the odds of getting A followed by C followed by G sequentially is equivalent to
0.25 x 0.25 x 0.25 = 0.015625 ... or 1 chance in 64 of getting it right.
As you can hopefully see by now, there's a simple formula in play, namely
0.25 ^ S where S represents the number of correct selections.
Here's a simple table:
As can be readily seen, the odds of nature randomly selecting the 1st 10 correct bases are at most approximately 1 million to one against ... which
over a long period of evolutionary time are darn good odds so no major problems so far.
But look what happens as we try to approximately double the number of correct bases from 10 to 25 ... the odds jump to a staggering 1 chance in
10^15.
And as we try for more and more correct bases, the odds go off the board when we try to get all 153 bases (insulin) with a truly mind bending odds of
1 chance in 10^92 !!!!
However, as TheWill pointed out, the insulin gene is a degenerate one, meaning that there are multiple ways of ending up still producing insulin but
with different arrangements of the 153 bases ... in fact there are approximately 10^24 alternative arrangements (per TheWills calculation) that still
result in insulin.
So using our original 10^92 calculation and reduce it by the 10^24 alternatives, we still end up with monstrously huge odds of approximately 10^68
against creating
any variant of the insulin gene.
To put it in perspective, 10^68 is the number 1 followed by 68 zero's !!!
Astyanax
Originally posted by Astyanax
reply to post by tauristercus
Thats exactly it in a nutshell ... mathematical odds ... and there's no escaping them or bypassing them.
Good grief, man – still holding out?
What part of 'selection' don't you understand? Nature didn't have to roll the same four dice over and over again. She selected the
combinations she 'wanted' in a cumulative process. Each roll of the dice increases the likelihood that the next roll will be the one she wants.
Evolution through natural selection isn't mathematically unlikely, it is inevitable.
You're kidding me ? C'mon ... you've got to be !
Each roll of the dice increases the likelihood that the next roll will be the one she wants.
Each roll of the die is completely and utterly
INDEPENDENT of any previous rolls if you're only interested in a single, one time outcome e.g.
A or C or G or T and don't care about what came before or after.
But if you're looking at multiple outcomes such as A followed by T followed by G followed by C followed by C followed by A, etc, etc, etc then every
previous outcome
MUST be taken into account in the final overall probability calculation.
In the case of building the insulin gene from scratch,
each and ever every roll of nature's 4 sided (A, C, G and T) die has
exactly the
same probability as any other roll ... but the
CUMULATIVE probability is based on taking every previous outcome into consideration.
That probability is exactly one chance in 4 or if you prefer, P = 0.25.
It also makes no difference if somehow nature managed to get say, the 1st 15 nucleotides sequentially in place and then nothing further happened for
10 million years. The odds of adding the 16th nucleotide will still be
0.25^16
against or if you prefer, approximate chances of 1 in 10^9 against.