Originally posted by Jorge
View Post
I have reread it. I still see no error in my evaluation of it. Maybe you can point it out, since you clearly think one exists.
To help you out, i'll rephrase my argument to try to be as clear as possible. From my reading, it appears Gauger is making a rookie mistake when it comes to evolution. She's looking at the presence of 2 closely related genes, and correctly inferring that they are duplicates. But from that, she assumes that since nylonase is a new function, it must be a new gene that's evolved in a copy of a previously functional one. In fact, her whole argument is based on that being correct, but it's not clear that it is.
What is correct is that both of these genes descended from a common ancestral gene through duplication. What was the ancestral gene function? How did the ancestral gene get there? There's no indication of any of this in Gauger's post. Could the ancestral version have arisen through a frame shift? Absolutely. The easiest way to check is to see if there are frame shifted open reading frames in both of the duplicates. I don't know how to get a copy of the nylonase plasmid, or i'd do that myself.
And, in any case, it's clear from the example i provided that there are lots of genes that have arisen by frame shifts.
Which brings me to Gauger's post 2. In it, she argues that it would be extremely rare to create an open reading frame (i.e. potential protein coding DNA sequence) through random mutations. Her argument is that, in creating random sequences, you only ended up with a 900 base pair open reading frame in a very low percentage of tries. (She didn't get a statistically significant number with 50/50 AT/CG bases, and got 57 in a million tries with 60% GC content).
So, handful of a million. Sounds like a really small number! Must be rare.
There's a different way of looking at this. Any single base change in a long DNA sequence is essentially like creating 900 new 900 base sequences, since you can view it as being in position 1, position 2... position 900. And that's just in one reading frame. There are 3 reading frames in any given sequence, since amino acids are encoded by sets of 3 base pairs. So that's 3 x 900, or 2,700. And then you have the option of reading the DNA from the opposite strand, which doubles the number to 5,400. We'll use 5,000 to simplify the math. Now, you only need 200 random mutations to get to the million random sequence mark, with its handful of open reading frames.
(This isn't entirely true, given that the probability of an open reading frame will depend on the precise sequence where the mutation occurs, but this is just for illustrative purposes, as you'll see.)
How often do you get 200 random mutations? Well, on average, each time the human genome is copied when a cell divides, you get a single new mutation somewhere in it. How often does a human cell divide? Well, we all started as single cells. The average adult human has somewhere in the neighborhood of 37 trillion cells. So, there were 16.5 trillion cell divisions to get those cells. And 8.25 trillion to get the previous generation of cells, and 4.125 trillion to get those, and so on. I'm not going to do the math; suffice it to say it took over 30 trillion cell divisions to make us.
So, remember when we were looking at 200 cell divisions involved in creating Gauger's probabilities? You can divide 30 trillion by 200, and you'd end up with 150 billion. Even if we say that there's an only 1 in a million chance of creating a new open reading frame, that means the average human body's cells could contain as many as 150 billion new open reading frames. That's not what i'd consider rare.
As i said, this isn't meant to be taken literally - the human genome has lots of sequences that are more or less prone to being mutated to create new open reading frames, so the real number would depend on analyzing its sequence to see how many of them could be changed at a single location and create a 900 base open reading frame. But i meant it to be illustrative. Granger points to the 57 in a million number she calculated, and says "that's small!" But life involves some pretty vast quantities—quantities that make a million seem like a rounding error. (For example, someone calculated that a bacteria that divided once every 20 minutes would have 4,553,481,496,843,251,613,696 progeny a day later.)
In other words, Gauger's numbers might be right, but she ignores the context of where those numbers matter entirely.
EDITS: fixed typos.
Comment