What is a gene?

lappel at eagle.wesleyan.edu lappel at eagle.wesleyan.edu
Tue Feb 15 17:22:41 EST 1994


In article <2jjffj$dp5 at news.mic.ucla.edu>, Ron Kagan <rkagan at ewald.mbi.ucla.edu> writes:
> Re: What is a gene?
(much deleted)
>>  I also think that the definition of
> the
>>physical boundaries of a gene in the statement above should be expanded
> to
>>include all proximal and distal regulatory regions (promotors,
> enhancers, etc)
>>involved in the transcription of that gene.  This would not include
> soluble
>>regulatory molecules - just the appropriate regions of DNA.
>>
>>Rick Willis 
>>University of Rochester Cancer Center
> 
> 
> I think that the correct term for any stretch of DNA/RNA which leads to a
> RNA/DNA product is *transcription unit*, not gene.  The reason that I
> include biological function in my definition of "gene" is that a
> definition should include the most essential characteristic(s) of the
> entity being defined; that is, the one(s) that best explains all (or
> most) of the other characteristics of the entity.
> 
> In the case of a gene, even without knowing about transcription or
> reverse transcription, we know that genes are units of heredity, are
> factors that determine characteristics (phenotype) of an organism,  are
> carriers of biological information, and are subject to the forces of
> natural selection.   A non-functional transcription product does none of
> these.  It is soon lost (in terms of evolutionary time) from the "gene
> pool" because there are no selective pressures to maintain it, and it
> carries no information.  In this regard, it is like a string of random
> letters: "xfrk", which does not constitute a word, even though words are
> composed of letters, because it conveys no meaning (information).
> 
> Aside from "transcription unit", there is a term for a segment of DNA
> that is recognizable as having once encoded a functional product, but no
> longer does: "pseudogene".
> 
> *It is only because a gene encodes a biological function that make it all
> of the things that we know a gene to be: a unit of information, heredity,
> and natural selection.*
> 
> Therefore, I stick to my original definition:
> 
> A gene is a segment of DNA/RNA that, when
> transcribed/reverse-transcribed, 
> leads to a biologically functional product.
> 
> 
> Ron Kagan
> rkagan at ewald.mbi.ucla.edu

	Not that I particularly like your definition, but I think it could be
improved by adding the word "correctly" between the first and second lines. 
That would get it closer to including the proper REGULATION of the
transcription unit, which is an indispensible part of the GENE.
	Evolution, for instance, would definitely notice if my hemoglobin was
trnascribed and translated into functional protein (i.e., able to carry oxygen)
but in my toenails instead of my blood cells!  That is what the commenter at
the top of this edited reply was getting at.
	And note that I include proper translation as well as transcription. 
Post-transcriptional regulation does exist: see Ruvkun & Ambros's recent papers
in Cell (I believe) about regulation in worm development by  5' UTRs 
(Untranslated regions).  Again: a definition that says only "function" is
insuffient.  Is the right protein in the wrong place not functional?
	My definition:
	a LOCUS is the place on a chromosome where a gene normally (barring
tranposons and the like) resides.
	an ALLELE is the particular version of information carried at that
locus in a given individual.
	a GENE is all of the DNA (or RNA if it's an RNA genome)at that 
locus necessary to get correctly regulated expression of the gene's function
-- thereby including promoters, enhancers, and any other cis-acting material,
but excluding trans-acting factors such as polymerases, txn factors, etc.
	but what is EXPRESSION?  it could be transcript, or protein, or even
just chromatin structure.  Anything that evolution can work upon, I suppose.
	So: a new definition to trash! go for it!

Laurel F. Appel     (LAPPEL at EAGLE.WESLEYAN.EDU)
Genetics Group
Biology Dept.
Wesleyan University
Middletown, CT 06459



More information about the Bioforum mailing list