Co-dominant vs additive inheritance; what is the latter and what is the difference?

E. Wijsman wijsman at
Tue Jan 27 11:56:43 EST 1998

The concept of co-dominance includes additive models.  If you are dealing
with a quantitative measure of the phenotype, then if the heterozygote has
a mean level which is the average of the homozygote means you have an
additive model.  The model in this case is also codominant since a
codominant model is one in which the heterozygote mean differs from both
the homozygote means (and is usually also restricted to fall between them
siince if the heterozygote mean is outside the homozygotes means we
usually call the model overdominant).  

Research Professor			1914 N 34th St., suite 209
Div. of Medical Genetics and		Seattle, WA   98103
Dept. Biostatistics 			(Note:  do not mention the
BOX 357720, University of Washington 	 Univ. of Washington, and
Seattle, WA   98195-7720		 USE THIS ADDRESS ONLY FOR
phone:  (206) 543-8987			 EXPRESS MAIL !!!)
fax:    (206) 616-1973			email:  wijsman at

On 27 Jan 1998, Arthur Osterop Ph.D. wrote:

> Dear netters,
> I am studying the occurrence of a gene allele polymorphism (A- and 
> B-allele) in relation to hypertrophy. I have analysed the allele 
> frequencies in relation to the hypertrophy data in three models.
> - a recessive model for the B-allele: AA+AB vs BB
> - a dominant model for the B-allele: AA vs AB+BB
> both models with a Student's t test, and
> - a co-dominant model: AA vs AB vs BB
> with ANOVA analysis.
> Now a reviewer of a journal in wich I want to publish my data states 
> that my co-dominant model actually is an assumption of additive 
> inheritance.
> Can somebody tell me what is meant by additive inheritance and in what 
> way does it differ from a co-dominant phenotype of alleles. As far as I 
> understand co-dominant means that the phenotype of the heterozygote 
> differs from both the homozygotes.
> A question concerning the statistics is whether or not one needs to 
> perform a Bonferroni correction, in this case of a factor three as three 
> different analysis have been performed on the 'same' data.
> I am looking forward for your responses as I have never been in 
> population genetics before let be statistics,
> With best regards,
> Arthur Osterop, PhD
> Erasmus University Rotterdam
> osterop at

More information about the Gen-link mailing list