theories of dominance

Toby Bradshaw toby at stein.u.washington.edu
Wed Mar 24 17:07:04 EST 1993


In article <1oqdd8INNnrf at shelley.u.washington.edu> bishopj at botany.washington.edu writes:
>In article <1onudrINNmi7 at shelley.u.washington.edu>  
>toby at stein.u.washington.edu (Toby Bradshaw) writes:
>> In article <1onpe4INNi7r at shelley.u.washington.edu>  
>>bishopj at botany.washington.edu writes:
>> 
>> [snip]
>> 
>>jb>Furthermore, among mutations arising in a population  
>>jb>recessivity is clearly the  
>>jb>rule.
>> 
>tb> I doubt this is true.  For visible mutations, maybe.  If it were 
>tb>_generally_ true, narrow sense heritability estimates for quantitative
>tb>traits would always be low, and that's not the case.
 
jb>Yes, I would definitely like to recant that statement.   
jb>Neutrality is the rule.  
   ^^^^^^^^^^^^^^^^^^^^^^

I would submit that neutrality is a frequent _assumption_.

jb>Most mutations which do have an  
jb>effect on enzyme activity are expected to decrease it.   
jb>Because of the buffering effect of enzyme networks, these  
jb>mutations may have little or no phenotypic effect, i.e.  
jb>they have no detectable effect on flux through the enzyme  
jb>system. When mutations occur which do increase catalytic  
jb>activity, they are expected to be undetectable, since they  
jb>will already be on the plateau of the flux-enzyme activity  
jb>curve.  This class of undetectable mutations would  
                         ^^^^^^^^^^^^
jb>contribute to the genetic variance in a quantitative trait.   
jb>When a mutation does have a detectable effect, it will  
jb>usually be associated with a large decrease in enzyme  
jb>activity, and will be recessive or partially so.   

I admit to feeling a little silly posting this when John and
I could walk to each others' lab in five minutes, so feel free
to hit "n".

Mutations at quantitative trait loci (QTLs) are not undetectable,
but they may be more difficult to detect than simple mendelian
loci.  QTLs _become_ mendelian loci when isolated in otherwise
isogenic backgrounds, and there is no fundamental difference
between a QTL and a mendelian locus.

It is easy to imagine that random mutagenesis is more likely
to produce a null allele than other alleles, but the effect of a 
null can as easily be additive as recessive.  The idea that
the hemizygous condition is equivalent to the homozygous state
for many or most enzyme systems would not stand a lot of scrutiny.
Elaborate mechanisms, such as dosage compensation, have evolved
to produce hemi/homo equivalence in special cases.  The fact that 
dosage compensation is necessary at all argues that simple dominance 
is an inadequate method of achieving homeostasis.  The phenotypes
of monosomics, available in many crops, would be of interest
here.  Perhaps someone familiar with them will chime in.

Toby Bradshaw                       |
Department of Biochemistry          |  Will make genetic linkage maps
and College of Forest Resources     |            for food.
University of Washington, Seattle   |
toby at u.washington.edu               |



More information about the Mol-evol mailing list