Nylon eating bacteria problem

[lucaspa]

Any idea what this previously 472 residue-long arginine-rich protein did? Was it involved in carboydrate hydrolysis?

And how could they tell whether 6-aminohexanoic acid linear oligomer hydrolase had any activity on non-nylon substrates if the bacteria could effectively use those substrates with other enzymes?

Second one first: you purify the enzyme and test it on the substrates. This is standard procedure to determine the substrate of any enzyme.

It turns out that the original sequence seems to have been NON-coding. IOW, later work showed that the sequence was not translated to a protein. Thomas needs to update the website.

 

[lucaspa]

Since his paper is referenced to by universities (Example: "A good example is the evolution of a Nylon-digesting enzyme in Flavobacterium sp. K172 (Thomas 2000). Nylon, a synthetic fiber, did not exist before 1937. A group of Japanese scientists discovered a bacterial mat in a factory pond, and found that the bacteria living there are able to digest Nylon. They replicated the process in the laboratory and discovered that the Nylon-digesting ability was caused by the appearance of an entirely new enzyme, nylonase, which was formed by a point mutation. Nylonase was not a changed version of an old enzyme; it cannot work on other substances.")

Notice there is no error here. It is talking about the ENZYME, not the bacteria. And yes, the ENZYME has no other substrate.

 

[lucaspa]

Prof, could you do me a favour? You said you had the original papers, right? Someone in a discussion of this popular bug on another board is asking how a frameshift mutation, which results in a whole new set of amino acids, can still result in a functional protein when there are so many hundreds of amino acids involved. I know the new enzyme is supposed to be shorter than the original, which I assume means that it hits a stop codon earlier, but does the paper say how far along the gene the frameshift actually occurs? I mean, are all 400 (or whatever) amino acids new, or just the last two or three before the stop codon?

In Ohno's paper, it looks like the insertion came 33 amino acids into the 427 amino acid original sequence. That means that ALL 392 amino acids of the new protein are novel. What the mutation did was make a new start sequence, drop the original 32 initial amino acids, and make a new 392 amino acid protein.

Does that answer the question?

 

[Fozzwald]

Notice there is no error here. It is talking about the ENZYME, not the bacteria. And yes, the ENZYME has no other substrate.

"They replicated the process in the laboratory and discovered that the Nylon-digesting ability was caused by the appearance of an entirely new enzyme, nylonase, which was formed by a point mutation."
Point mutation or frameshift mutation?

 

[Cantuar]

Prof, yes it does, thanks. I know the next question this guy's going to come up with, though, which will be along the lines of the massive improbability of such a mutation, so early in the sequence, affecting so many codons and still resulting in a workable protein rather than a total meltdown. Is there anything about frameshifts that would make such a scenario more likely than it appears to be on the surface?

 

[lucaspa]

"They replicated the process in the laboratory and discovered that the Nylon-digesting ability was caused by the appearance of an entirely new enzyme, nylonase, which was formed by a point mutation."
Point mutation or frameshift mutation?

Yes.

A frameshift mutation is one type of point mutation. There are 3 types of point mutation:

1. Substitution
2. Insertion
3. Deletion.

The last two result in a "frameshift" because there is no punctuation in the genetic code. Every 3 bases is a "codon" and is translated to an amino acid. Take one base out or put one in and the "reading frame" is shifted so that you get different groups of 3 bases.

For example:

ATCGGCTACCCA breaks down to ATC, GGC, TAC, CCA

Now, insert a T in the front so you have ATTCGGCTACCCA and now you have the groups ATT, CGG, CTA, CCC

Still a point mutation but results in a frame shift.

Is that clear?

 

[lucaspa]

Prof, yes it does, thanks. I know the next question this guy's going to come up with, though, which will be along the lines of the massive improbability of such a mutation, so early in the sequence, affecting so many codons and still resulting in a workable protein rather than a total meltdown. Is there anything about frameshifts that would make such a scenario more likely than it appears to be on the surface?

1. There are thousands or millions of 397 amino acid proteins that would hydrolyze nylon. In fact, notice that the plasmid now has 2 proteins that do so. A gene duplication and then modification of the second gene that gives it a 160 fold greater activity than the original enzyme. So, there is not just ONE 6-aminohexanoic acid dimer hydrolase, but millions of possible ones.

2. Think of this as the lottery. Yes, the odds of any particular individual ticket matching the 6 numbers is 1 in millions, when you have 200 million tickets, it becomes virtually certain that ONE of the tickets will match.

So, consider each bacterium a ticket. There are billions of bacteria in just a liter of water. Each of those bacteria divides every 20 minutes or so. So in just one year you have quadrillions of lottery tickets -- bacterium. No matter how improbable it may be for a particular bacterium to get the mutation, it now becomes certain that ONE bacterium will mutate to ONE of the possible hydrolases.

That is something that creationists have trouble grasping. For the individual, it may be improbable that YOU will have the mutation. But for the POPULATION, it is certain that SOMEONE will be lucky. For the individual, it is improbable that you will be the lucky one. And creationists hate that lack of control. There is nothing the individual can do to influence its genes. That was decided for humans at the moment of conception; for the bacterium at the moment of division. We can't do a thing to influence our evolutionary fitness. It is all handed to us in our genes. Nothing, not hard work, prayer, belief, NOTHING, will make our genes the "chosen" genes.

It's like straight poker. All you can do is play the hand you are dealt. No drawing new cards or discarding bad ones. You either have a winning hand or you don't.

 

[Fozzwald]

1. There are thousands or millions of 397 amino acid proteins that would hydrolyze nylon. In fact, notice that the plasmid now has 2 proteins that do so. A gene duplication and then modification of the second gene that gives it a 160 fold greater activity than the original enzyme. So, there is not just ONE 6-aminohexanoic acid dimer hydrolase, but millions of possible ones.

2. Think of this as the lottery. Yes, the odds of any particular individual ticket matching the 6 numbers is 1 in millions, when you have 200 million tickets, it becomes virtually certain that ONE of the tickets will match.

So, consider each bacterium a ticket. There are billions of bacteria in just a liter of water. Each of those bacteria divides every 20 minutes or so. So in just one year you have quadrillions of lottery tickets -- bacterium. No matter how improbable it may be for a particular bacterium to get the mutation, it now becomes certain that ONE bacterium will mutate to ONE of the possible hydrolases.

That is something that creationists have trouble grasping. For the individual, it may be improbable that YOU will have the mutation. But for the POPULATION, it is certain that SOMEONE will be lucky.

It's only certain if the events are dependent on each other. Using your lottery analogy...say the true odds of a lottery are 1 million to 1. If I have 1 million tickets to buy I can guarantee winning because I know what sets of combinations I have played and I can cover all the possibilities, so each set of new numbers I play is dependent on the previous ones I played. But if I were to play 1 ticket a week for the next million weeks that guarantee isn't there because each drawing is independent of the other. If I were to play 10 tickets a week for a million weeks the same holds true. If I were to ask the computer to pick my million numbers I still wouldn't be certain of winning. Even if the computer printed 2 million tickets for me, I'm not positive of winning.

If 1 million of my closest friends were each to buy one ticket in a given week we wouldn't be guaranteed of hitting the lottery if we didn't compare numbers and cover all possibilities (dependence). The same holds true if my 1 million friends were to buy one ticket a week for a million weeks. None of us would be guaranteed winners.

If an occurance is deemed mathmatically impossible, it will not happen no matter how many times it is attempted if the attempts are independent of each other, even if the number of attempts exceeds the odds. Say an event is deemed as possible once in 5 (100 to the 200th power). If I make each attempt independent of each other it won't happen even if I attempt it 20 (100 to the 200th power) times. That's why all of the "given enough time the impossible becomes improbable and the improbable becomes likely" arguments are in error - a lack of understanding probabilities and odds. (I knew all those gambling books I read would come in handy someday.)

 

[Arikay]

I think you missed the point.

Like take the lottery example. someone often wins the lottery. What are the chances it was you? Very unlikely. What are the chances someone, somewhere won? Much more likely.

The bacteria as a population, are just like people playing the lottery. What are the chances a single bacteria will "win" the lottery and get the mutation? Very small. What are the chances that some bacteria, somewhere will "win?" Much more likely.

Unless of course you are saying that its improbably for someone, somewhere to win the lottery, in that case, im sure the winners would tell you otherwise.

It's only certain if the events are dependent on each other. Using your lottery analogy...say the true odds of a lottery are 1 million to 1. If I have 1 million tickets to buy I can guarantee winning because I know what sets of combinations I have played and I can cover all the possibilities, so each set of new numbers I play is dependent on the previous ones I played. But if I were to play 1 ticket a week for the next million weeks that guarantee isn't there because each drawing is independent of the other. If I were to play 10 tickets a week for a million weeks the same holds true. If I were to ask the computer to pick my million numbers I still wouldn't be certain of winning. Even if the computer printed 2 million tickets for me, I'm not positive of winning.

If 1 million of my closest friends were each to buy one ticket in a given week we wouldn't be guaranteed of hitting the lottery if we didn't compare numbers and cover all possibilities (dependence). The same holds true if my 1 million friends were to buy one ticket a week for a million weeks. None of us would be guaranteed winners.

If an occurance is deemed mathmatically impossible, it will not happen no matter how many times it is attempted if the attempts are independent of each other, even if the number of attempts exceeds the odds. Say an event is deemed as possible once in 5 (100 to the 200th power). If I make each attempt independent of each other it won't happen even if I attempt it 20 (100 to the 200th power) times. That's why all of the "given enough time the impossible becomes improbable and the improbable becomes likely" arguments are in error - a lack of understanding probabilities and odds. (I knew all those gambling books I read would come in handy someday.)

 

[DrLao]

But regardless of claims that it is mathematically impossible, it did happen. Therefore, the model used to judge it mathematically impossible must be flawed.

 

[Cantuar]

Yes, but this is probably where we get to hear that since it's mathematically impossible to occur by chance but it did happen, then it must have been God. The sound of yet another strawman being slain.

 

[Jet Black]

yeap... having looked around at the evidence, there is just too much that points to at the very least, the possibility of spontaneous life... too many coincidences such as the fact that the majority of amino acids in asteroids are left handed, and so on....

 

[lucaspa]

It's only certain if the events are dependent on each other. Using your lottery analogy...say the true odds of a lottery are 1 million to 1. If I have 1 million tickets to buy I can guarantee winning because I know what sets of combinations I have played and I can cover all the possibilities, so each set of new numbers I play is dependent on the previous ones I played. But if I were to play 1 ticket a week for the next million weeks that guarantee isn't there because each drawing is independent of the other. If I were to play 10 tickets a week for a million weeks the same holds true. If I were to ask the computer to pick my million numbers I still wouldn't be certain of winning. Even if the computer printed 2 million tickets for me, I'm not positive of winning.

You changed the parameters. Instead of talking about a winning ticket, you talked about a winning INDIVIDUAL. Again, you are thinking of the individual and not the population. In a lottery you have millions of tickets printed and each is independent. Right? So many of the tickets could duplicate. However, once you get 2x the odds, you reach close to unity that ONE of those tickets will have the numbers.

Same way with proteins.

If an occurance is deemed mathmatically impossible, it will not happen no matter how many times it is attempted if the attempts are independent of each other, even if the number of attempts exceeds the odds. Say an event is deemed as possible once in 5 (100 to the 200th power).

That may be true, but it isn't relevant here. We don't have odds nearly that high. And we have independent "tickets" well over 2x the odds.

 

[Cantuar]

OK, ignore what I said here originally. Getting confused again...

 

[Happy Wonderer]

If an occurance is deemed mathmatically impossible, it will not happen no matter how many times it is attempted if the attempts are independent of each other, even if the number of attempts exceeds the odds. Say an event is deemed as possible once in 5 (100 to the 200th power). If I make each attempt independent of each other it won't happen even if I attempt it 20 (100 to the 200th power) times. That's why all of the "given enough time the impossible becomes improbable and the improbable becomes likely" arguments are in error - a lack of understanding probabilities and odds. (I knew all those gambling books I read would come in handy someday.)

I think you don't completely understand the concept of statistical independence or dependence. Yes, if something is strictly impossible (probability = zero) then no matter how many trials you have it will not happen. Zero times anything to the power of anything is still zero. If the probability (P) is greater than zero, then the number of trials affects the outcome.

Arrgh, I don't know how to do equations in HTML, look up "binomial distribution" to see how to calculate the odds that an event of probability P will occur at least once in N trials. You will see from the formula that as N approaches infinity, the the probability of of the event occurring at least once approaches unity. The odds of winning on any given trial are the same. (Think about it -- what are the odds of not flipping heads in 20 coin flips. Is it more or less than the odds of heads on the first flip?)

Your lottery example of covering all the odds is different than what anyone discussing the probability of a beneficial mutation is claiming. You are increasing the likelyhood of success in an individual trial by changing P. If you bet on both heads and tails of a coin flip, you will always win. This has nothing to do with independence. An example of dependent (actually 'contingent') probability would be the odds of pulling a red jelly bean out of the jar after you have already pulled six blue ones from it. (Your example would be equivilant to deciding that either a red or a black jelly bean was a 'win'.)

hw

 

[Fozzwald]

I think you missed the point.

Like take the lottery example. someone often wins the lottery. What are the chances it was you? Very unlikely. What are the chances someone, somewhere won? Much more likely.

Much more likely, yes. Certain, no. But "certain" was the word that was used.

 

[lucaspa]

Like take the lottery example. someone often wins the lottery. What are the chances it was you? Very unlikely. What are the chances someone, somewhere won? Much more likely.

Much more likely, yes. Certain, no. But "certain" was the word that was used.

If there are enough tickets, it becomes certain that someone will win. See the post above yours.

Happy Wonderer "Arrgh, I don't know how to do equations in HTML, look up "binomial distribution" to see how to calculate the odds that an event of probability P will occur at least once in N trials. You will see from the formula that as N approaches infinity, the the probability of of the event occurring at least once approaches unity."

Unity = certain.

Now, if you want to nitpick over "certain" you can. However, a probability of 0.99999 may not be absolutely certain, but it is more than close enough for the counter to your original argument.