🚀 go-pugleaf

P J Kennedy (paulk@socs.uts.EDU.AU) wrote:

: Hi all,

: I am a computer scientist trying to understand genetics. With this
: in mind, I have the following questions.

: I understand the term locus to mean the position of a gene on a chromosome.
: For example, if a gene spans the 100th to the 175th base pairs of the
: chromosome then its locus is 100.

: I understand the term allele to mean genes which have different effects
: on a single characteristic that are at the same loci on homologous 
: chromosomes.

I think you will find things less confusing if you avoid the word
"gene" in defining allele -- gene has too many overlapping meanings.

To use a computer analogy:
	Think of a chromosome as a large segment of memory
	Think of a locus as a contiguous range of addresses
	Once you have defined a locus, then each possible
         set of values filling the locus defines an allele.
         In real practice, we can only distinguish alleles if
         we have an assay for them -- some functional characteristic
         classically, but with molecular techniques it is possible
         (though not necessarily useful) to distinguish any two alleles.

: Is the actual locus of a gene on the genetic strip important ?
: From what I understand I don't think it is because a gene will either always
: be expressed (ie. it's constitutive) or it will be turned on or off by other
: genes and this is controlled by what the gene or its promoter looks
: like not its actual position on the strip.

For most genes, the position of the locus on the chromosome is not important
when determining the phenotype generated by a given genotype (set of alleles).

: So if the position isn't important why should alleles be in the same
: position on different chromosomes ? 


: Why should they even be the same length ?
: (Indeed, are they the same length ?) 

This is actually a significant question -- not all alleles of a locus are
necessarily the same length.  Two common classes of null (non-functional)
alleles are those due to the insertion or deletion of genetic material.
However, for most analyses the precise length of the locus isn't important.
I think it is important to remember that locus and allele are abstract
concepts that don't always map perfectly onto biological reality.
For example, a deletion may completely remove a given locus.  For
some types of analysis, we treat this as a null allele at that locus,
whereas for others it is important to treat this as the locus disappearing.
To give another example, an additional means to generate a null allele
is if the locus is the breakpoint of a genetic rearrangement.
If we held strictly to my address analogy, the address range (locus) has
now been subdivided into two non-contiguous segments.  Has one locus
become two?  Well, for most analyses, no.  We just have a null allele.

: Is crossover the mechanism that forces
: genes with similar effects to reside at the same locus ? ie. does crossover
: force alleles ?

Replace "genes" with "alleles".  In one sense, by definition the two
alleles must reside at the same locus (otherwise they aren't alleles).
What you are probably more interested in is the question as to why
genes are organized into consistent chromosomes, 
and its not so much crossing-over
(though crossing-over probably wouldn't occur if they weren't) as
simply the inherent result of how chromosomes are made -- by duplicating
other chromosomes.  

: If, by inversion, an allele of a gene is moved from one locus to another
: position, is it still an allele ? If it is, then the whole definition of
: alleles seems, to me, to be inconsistent.

Again, the inversion moves the locus physically, but in the abstract
sense you still have the same locus (it has the same responsibility).
You might think of genetic loci as defining "genetic space", and 
physical locations (chromosomes) defining "chromosome space".  Alleles
are defined and loci delineated at a time when 
genetic space and chromosome space are interchangable.
Deletions, insertions, rearrangements, duplications, etc break
the interchangability of the two spaces, but the definitions of the
alleles and loci remain fixed.

: Is the garbage information that now fills its old locus an allele ? If it
: isn't (as I would expect) then does that mean that at a particular locus one
: finds either alleles of a gene, garbage or other genes ? 

: Is it right to say that base pairs in the promoter region of a gene can
: serve as some kind of signature for that gene ? For example, if an enzyme
: wants to repress a gene for short hair and the promoter region for the
: short hair gene always contains the sequence AAATTGGGTACG then can the
: enzyme only look for this sequence rather than a much longer one that is
: the actual gene itself ?

Yes.  Transcription factors recognize short sequences, not entire genes.


: Must each of the alleles of a particular gene have the same information
: in the promoter sequence ? By that I mean, do all alleles of a gene have to
: be turned on in the same way and have the same signature ?

No.  Many alleles are defined by alterations in the promoter regions.
These can either eliminate expression (another way to make a null
allele), cause reduced expression, or 
cause inappropriate ("ectopic") expression.



Keith Robison
Harvard University
Department of Cellular and Developmental Biology
Department of Genetics / HHMI

robison@mito.harvard.edu