molecular drive

Joe Felsenstein joe at GENETICS.WASHINGTON.EDU
Fri Nov 8 14:10:34 EST 1991

Mark Yandell expressed surprise that

> Does anybody out there know anything about molecular drive?
> To the extent that I understand his argument, homoginization 
> results in a small variance for the trait in the population
> regardless of the frequency of the trait(mutation) in the
> population and then(HERE IS WHere I AM MOST CONFUSED)
> the particulars of the population such as mating structure
> ,etc. act to fix the mutation in the ABSENCE OF BOTH SELECTION

Gabriel Dover gave the name "molecular drive" to a series of processes.
He argued that it was a newly discovered evolutionary force.  One of
the central processes was the already-known "concerted evolution" that occurs
when unequal crossing-over and gene conversion spread the same mutations
through a gene family.  In the unbiased conversion case Yandell was rightly
puzzled about,

(1) Different copies in the multigene family in the same chromosome or
gamete come to be very similar, but

(2) Different chromosome copies could carry different alleles: such as
    A1-A1-A1-A1-A1 in one and   A2-A2-A2-A2-A2 in another.
 In fact, homogenisation within a chromosome happens but there is no
 homogenisation of the population.  In fact in such cases the population
 is MORE variable than if there were only one copy in the gene family.
 (I can explain this if needed).

(3) The rate of substitution at any one site is EXACTLY the same as if
 there were no multigene family!  This fixation is due to drift, as usual.
 There is thus NO extra evolutionary force in this case.  The only difference
 is that the alleles fixing at different copies in the family have a very
 strong tendency to be the same alleles.

I conclude that molecular drive is a misnomer and that one understands the
process better if it is called concerted evolution.

Joe Felsenstein, Dept. of Genetics, Univ. of Washington, Seattle, WA 98195
 Internet:         joe at     (IP No.
 Bitnet/EARN:      felsenst at uwavm
 UUCP:             ... uw-beaver!evolution.genetics!joe

More information about the Mol-evol mailing list