Re: Haldane's Dilemma - clarifications - and Felsenstein [LONG]


"Joe Felsenstein" <joe@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message news:e80qcc$2egr$1@xxxxxxxxxxxxxxxxxxxxxx
In article <e7vngj$1vbp$1@xxxxxxxxxxxxxxxxxxx>,
Perplexed in Peoria <jimmenegay@xxxxxxxxxxxxx> wrote:
Actually, I believe that unsuccessful substitutions are a feature
of this simulation. Malcolm commented that (in some regimes) he
was seeing that the proportion of beneficial alleles fixed was
1/2s, which Felsenstein then identified as theoretically justified
by Haldane, and later, more accurately, by Kimura. So, a fraction
(2s-1)/2s of the attempted substitutions were unsuccessful.

I think the figure was 2s, not 1/(2s). The latter is not a fraction,
in cases where s is small, as would usually be the case. For example
when s = 0.01, the probability of fixation is about 0.02 and the
probability of loss is 0.98.

Dooohh! Right.

Deleterious mutations could be added to Malcolm's simulation if needed
(along the lines I suggested). I doubt they would make much difference
if the number of advantageous mutations was still about the same, as almost
all of the deleterious mutations would be rapidly lost.

Of course ReMine believes that the 'cost' of removing these deleterious
mutations might be significant and might therefore lower the limit.
Plausible enough, but I wouldn't think that it would be a significant
factor. The cost of fixing an advantageous mutation is the cost of
incrementing the count 2N-1 times. The cost of removing a deleterious
mutation is the cost of decrementing the count once. So I agree with
you that it probably wouldn't make much difference. But ReMine makes
noises about different impacts for recessive vs dominant deleterious
mutations. He may be right. I haven't thought about it enough.


.

Relevant Pages

  • Re: Reproductive Excess: Is Required
    ... >>> for all selection coefficients). ... >>> cost is 999,999, which is a huge cost. ... We have speeded up substitution by a factor of 300, ... what we have is a series of events that first fix pre-existing mutations ...
    (sci.bio.evolution)
  • Re: The cost of substitution
    ... mutations in fact have a substantial cost of substitution -- and Dr. ... Warren Ewens emphatically acknowledges I am CORRECT. ...
    (sci.bio.evolution)
  • Re: Haldanes Dilemma - clarifications - and Felsenstein [LONG]
    ... Your simulation goes beyond that, to make even more wildly unrealistic ... In other words, your simulation begins with ONLY BENEFICIAL mutations, ... and drastically reduces the cost of mutation (the cost associated ...
    (sci.bio.evolution)
  • Re: The cost of substitution
    ... > insists, as ReMine implicitly does, that some huge number of ... > mutations would have been required to get a 'sapien from a simian', ... CLEAR cost of substitution argument DOES MATTER to any valid ...
    (sci.bio.evolution)