Register | Sign In


Understanding through Discussion


EvC Forum active members: 58 (9173 total)
3 online now:
Newest Member: Neptune7
Post Volume: Total: 917,557 Year: 4,814/9,624 Month: 162/427 Week: 75/85 Day: 12/12 Hour: 2/0


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Evolutionists improbable becoming probable argument
Percy
Member
Posts: 22604
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


(2)
Message 4 of 98 (907207)
02-20-2023 10:13 AM
Reply to: Message 1 by mike the wiz
02-19-2023 8:12 AM


mike the wiz writes:
Evolutionists have classically argued that if something is not a matter of being intrinsically impossible that therefore it's a matter of probability and that no matter how improbable eventually the event will become probable.
I recognize the issue you're trying to describe, but as nwr said, that's not really what they're saying. For one thing, they're not over-generalizing the way you are to anything improbable.
But there are certain events in life that while individually improbable become probable under normal circumstances. For example, if last year an omicron virus needed a certain mutation in order that its spike protein better evade detection by the immune systems of humans, what are the odds that it will experience it? Very tiny. Let's call the odds one in a billion, which is .000000001 or 10-9.
But an infected person has between a billion (109) and a hundred billion (1011) covid viruses in his body, and at any given instant in time there's around a million covid cases worldwide (106), that would put the total number of covid viruses worldwide at any given instant in time at a minimum of 109 × 106 = 1015 and a maximum of 1011 × 106 = 1017.
Recall that assigned odds of this specific mutation occurring in a single covid virus at 10-9, so given the huge number of covid viruses in the world at any time, the odds of this mutation occurring in at least one of them are a minimum of 1-(1 - (10-9)1015 = .9999... or effectively 1. The maximum is a decimal point followed by even more 9's, also effectively 1. The odds of this mutation occurring in at least one covid virus are a virtual certainty.
If you make the odds of the mutation even less likely, say one in a trillion or 10-12 then the answer doesn't change. It's still effectively 1. There are just so many covid viruses worldwide that beneficial mutations (for them) become a virtual certainty.
If you make the odds of this mutation small enough, say one in a quadrillion or 10-15 then the odds finally become less than effectively 1, in this case .63.
I'm feeling particularly rusty at probability today. Someone might want to check my math.
You mentioned abiogenesis several times (and confused it with evolution yet again). What step or steps in the process of abiogenesis do you consider particularly unlikely?
--Percy

This message is a reply to:
 Message 1 by mike the wiz, posted 02-19-2023 8:12 AM mike the wiz has not replied

  
Percy
Member
Posts: 22604
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


(2)
Message 27 of 98 (907759)
02-28-2023 9:42 AM
Reply to: Message 14 by sensei
02-27-2023 7:55 PM


sensei writes:
For a gene of 100 bases, there are already more than 10^60 possible sequences. How many of these would be usefull?
It's not possible to answer the question, "How many of the possible sequences would be neutral or better than neutral compared to the current sequence?" And any answer would change over time because of changing circumstances at the protein, phenotype and environmental levels.
And this is for the low count. For 200 bases, the number of possible sequences is already more than the estimated number of atoms in the entire universe.
Because it affects the probabilities you're calculating quite substantially it's worth noting that nucleotides are grouped into codons. 61 specify amino acids and the remaining 3 are start/stop codons. But there are only 20 amino acids, so there is some redundancy in that more than one codon can represent the same amino acid.
So the probability isn't what you calculated since what matters isn't the nucleotide sequence but the amino acid sequence that the codons correspond to. Your calculation is 4198 = 1.6 × 10119 (198 instead of 200 so that it is a multiple of 3 making 66 codons) is the number of random nucleotide sequences. But the beginning and end of the gene's codon sequence have to be start and stop codons, so that means only 64 coding codons. And since there are only 20 amino acids, the actual probability of any given random codon sequence is 2064 = 1.8 × 1083, nearly 40 orders of magnitude smaller than your original calculation.
For larger genes, the space grows exponentially. Do you think the number of useful sequences grows just as fast? Very unlikely, I'd say.
Why do you think it very unlikely? How would you know? Isn't it a function of influences independent of the sequence itself and inherently unknowable?
But obviously the answers are very favorable to life. SARS-Cov-2 is a recent example, evolving merrily around our defensive strategies.
--Percy

This message is a reply to:
 Message 14 by sensei, posted 02-27-2023 7:55 PM sensei has replied

Replies to this message:
 Message 42 by sensei, posted 02-28-2023 1:11 PM Percy has replied

  
Percy
Member
Posts: 22604
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


(1)
Message 50 of 98 (907820)
02-28-2023 4:37 PM
Reply to: Message 42 by sensei
02-28-2023 1:11 PM


sensei writes:
If you put random characters on paper, with five characters, it may happen once so often that you get a meaningful word or short sentence. With twenty characters, it's generally less likely to get a full meaningful sentence.
Yes, of course.
Do you have any reason or evidence to support the idea of more useful sequences at greater lengths, compared to the total number of possible arrangements?
It's self-evidently true that the greater the number of permutations the least likely any particular one is, including ones that make sense in English, but from following the discussion it seems that everyone understands this already. The difference of opinion lies in how to properly apply this information.
I think your main point is that it's more likely for a change to be in a deleterious direction than beneficial. In essence you're arguing that the sheer unlikelihood of a beneficial mutation means that evolution isn't possible.
But as I pointed out to mike the wiz in Message 4, there are around 1016 SARS-Cov-2 viruses in the world at any given point in time, and they replicate approximately every 10 hours. With 2.4 × 1016 replications per day and 10-6 mutations per replication, that's 2.4 × 1010 mutations daily. If the odds of a beneficial mutation are 10-9 (one in a billion) then that's in the neighborhood of 24 beneficial mutations every day. Day after day. With viruses with beneficial mutations being selected for.
--Percy

This message is a reply to:
 Message 42 by sensei, posted 02-28-2023 1:11 PM sensei has replied

Replies to this message:
 Message 53 by sensei, posted 03-01-2023 3:46 AM Percy has replied

  
Percy
Member
Posts: 22604
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


(1)
Message 59 of 98 (907857)
03-01-2023 8:54 AM
Reply to: Message 53 by sensei
03-01-2023 3:46 AM


sensei writes:
Well, it does not appear to be self-evident to Taq. But he thinks he understands probability better. Little does he know.
I think you're placing greater trust in errors in words and sentences as an accurate analogy to mutations in a gene than anyone else here. The analogy might serve as a helpful introduction to the concept of mutations but shouldn't be carried too far. Taq has experience as a molecular biologist, and he echoed my thought (or maybe I echoed his, I forget who said it first) that longer genes might be more tolerant of random mutations than short ones.
Here's an example. Take the sentence, "I found a watch," and assume it's an analogy to a gene with the nucleotide sequence "CAG...TAC" Now introduce a random mutation so that the sentence becomes, "I found a wmtch," and that analogically the nucleotide sequence becomes "CAG...AAC". In your opinion is the sentence destroyed, or would most people make the right supposition anyway. I certainly don't know. And for the nucleotide sequence where TAC becomes AAC, in terms of what the protein does in the organism does it matter that the last amino acid is now asparagine instead of tyrosine? I don't know that either.
But you are in effect claiming you know while at the same time are apparently unable to communicate that knowledge to anyone else. Hence the skepticism everyone's expressing. At least in this sub-discussion with me, your belief that errors in words and sentences are an accurate analogy to the occurrence and selection of mutations is a barrier to developing a common understanding. The analogy could be improved if you added the creation of copies of the sentence and a selection mechanism.
What we do know is that nearly every reproductive event includes mutations. Reproduction is imperfect. Someone mentioned that each human averages about 50 mutations. I thought the figure was around 100, but 50 is fine, too. The key point is that despite the mutations the human race continues. It is, in fact, evolving faster than ever, because the more babies born every year the more mutations introduced into the world population. It's also worth pointing out that zygotes and fetuses with sufficiently deleterious mutations are often spontaneously aborted, so early that many women never suspect they were pregnant.
Writing about this made me curious about what percent of the world population has genetic disorders. This is according to Genetic disorder - Wikipedia:
Wikipedia:
Around 65% of people have some kind of health problem as a result of congenital genetic mutations.
That's much higher than I would have suspected, and if this is accurate then the majority of these health problems must be minor. The article also says that about 600 genetic orders are treatable. For example, LVH (left ventricular hypertrophy - thickened heart muscle around the left ventricle that prevents the heart from efficiently pumping high volumes) has a strong genetic component but little practical effect (in most cases) unless you want to become an elite athlete, and it's treatable with medication. Are conditions like that included in the 65%? I suspect that they are including minor conditions like this. The prevalence of life-impacting genetic disorders must be much less, and looking up just one, muscular dystrophy, for example, I see that it occurs at a .03% rate.
Realize that if it were true that deleterious mutations win out over neutral and beneficial ones that all life would have come to an end long ago. You have to keep in mind that evolution includes mutation *and* selection. Disadvantageous mutations are selected against and tend to be removed from populations or at least become uncommon, while beneficial mutations spread rapidly through populations, and if they're sufficiently beneficial then their alleles becomes fixed (means every individual has them).
--Percy

This message is a reply to:
 Message 53 by sensei, posted 03-01-2023 3:46 AM sensei has not replied

  
Percy
Member
Posts: 22604
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


(3)
Message 76 of 98 (907922)
03-02-2023 9:03 AM
Reply to: Message 62 by sensei
03-01-2023 1:07 PM


sensei writes:
Nobody is talking about a single sequence here. That is a straw man and you know it.
You might be misinterpreting where Taq agreed with you that "the longer a sequence is the less probable any single sequence is." This is all part of the sharpshooter analogy. To briefly explain it again, the new nucleotide sequence resulting from a beneficial mutation is highly unlikely, but it is no more unlikely than the original sequence.
The sharpshooter analogy comes into play when someone claims that only a specific beneficial mutation, i.e., a specific sequence in a specific gene, is necessary for improved adaptation in the existing environment, but that's not true. There are thousands of genes and hundreds or thousands or millions or billions or trillions of reproductive events every day, depending upon the reproductive rate of the organism in question. Almost every reproductive event includes random mutations, and some tiny percentage of them will produce improved adaptation.
But a tiny, tiny percentage of a huge, huge number is still a healthy number. For example, about 400,000 human babies are born every day, and on average each has about 50 mutations, mostly SNP's. That's a total of 2 million random human mutations every day. What percentage of all possible mutations would produce improved adaptation? I don't know, but let's say that a beneficial mutation has only a 1 in a billion chance, or 1 in 10-9. This means that the likelihood of a beneficial mutation in a human baby on any given day is 1 - (1-10-9)(2 × 106) = .2%. There are 365 days in a year, so the probability of a beneficial mutation in any given year is 1 - (1-.002)365 = .52. So there's a 52% probability of a beneficial mutation somewhere in the human population every year.
And that's using a probability of 1 in a billion for a beneficial mutation. According to Population Genetics Made Simple the probability is much higher than that.
Funny how you always work towards nested hierarchy. Biased much?
A nested hierarchy is what is observed in nature. Accepting observations of the real world isn't bias.
--Percy

This message is a reply to:
 Message 62 by sensei, posted 03-01-2023 1:07 PM sensei has not replied

Replies to this message:
 Message 77 by Parasomnium, posted 03-04-2023 4:10 AM Percy has replied

  
Percy
Member
Posts: 22604
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


Message 79 of 98 (907975)
03-04-2023 9:23 AM
Reply to: Message 77 by Parasomnium
03-04-2023 4:10 AM


Parasomnium writes:
Percy writes:
That's a total of 2 million random human mutations every day.
Percy, not to detract from your argument, but you're off by a factor of 10, because 400,000 x 50 = 20 million, not 2 million. If you do the same calculation with the correct number your result of 52% shoots up to 99.9%. So the maths is even more in your corner than you thought.
Oh, right. Here's that paragraph again with correct figures:
But a tiny, tiny percentage of a huge, huge number is still a healthy number. For example, about 400,000 human babies are born every day, and on average each has about 50 mutations, mostly SNP's. That's a total of 20 million random human mutations every day. What percentage of all possible mutations would produce improved adaptation? I don't know, but let's say that a beneficial mutation has only a 1 in a billion chance, or 1 in 10-9. This means that the likelihood of a beneficial mutation in a human baby on any given day is 1 - (1-10-9)(2 × 107) = 1.98%. There are 365 days in a year, so the probability of a beneficial mutation in any given year is 1 - (1-.0198)365 = .999. So there's a virtual certainty of at least one beneficial mutation somewhere in the human population every year.
Thanks for the correction.
--Percy

This message is a reply to:
 Message 77 by Parasomnium, posted 03-04-2023 4:10 AM Parasomnium has replied

Replies to this message:
 Message 80 by Parasomnium, posted 03-04-2023 9:34 AM Percy has seen this message but not replied

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024