In message .com,
Raphanus writes
Give that alleles are different codings of the same gene - does that
mean the different alleles are the same length? It must also be true -
right? - that different alleles encode for different proteins (else
there would be no difference in their manifestation.)

Different alleles are often, but not universally, of the same length. In
some cases they are distinguished by one or more base pair
substitutions. In others they are distinguished by the insertion or
deletion of lengths of DNA.

Alleles don't always code for proteins - apart from ribosomal and
transfer RNA genes there are also bits of DNA which represent regulatory
sequences rather than protein, and other odds and ends.

It is possible for one allele to represent the absence of the gene; in
the case it obviously doesn't code for a program. In other cases the
protein product may be the same, but modifications to regulation have
changed the amount produced.

Assuming two samples of DNA could've - but not necessarily - come from
different species, if at a particular locus the DNA is different, how
does one know if the difference distinguishs different alleles or
different genes? .

With difficulty. If the genes are orthologous and functional then the
amount of difference between the genes (for that particular locus) is a
strong hint as to whether they come from the same or different species.
However the existence of paralogs and pseudogenes means that this
assumption cannot be relied on.

[Orthologs are the "same" gene in different lineages; paralogs are genes
which descend from different branches of a gene duplication event. For
example haemoglobin alpha and haemoglobin beta are paralogs; human
haemoglobin alpha and chimpanzee haemoglobin alpha are orthologs (and,
IIRC, in this case, identical).]

Thanks.


--
Stewart Robert Hinsley