----- Original Message -----
Sent: Tuesday, January 20, 2004 8:05
PM
Subject: [SpaceAgeRobin] Using irises
with SA in their genes............
Kelly, if you have a copy of TWOI
there is a section on tetraploid genetics. If you don't have it in your
library, it is very worthwhile to get it. The AIS Storefront has it on
sale at $15 I think.
Figuring the "Plunnett square" on
a tetraploid cross is a *lot* more complicated, not only because there are
four rather than two chromosomes involved, but because the manner of pairing
of the chromosomes affects the outcome.
In simple, straightforward
diploids (that's a laugh!--genetics, even in diploids, is never simple)
chromosomes either pair two by two, or not at all, depending on the hybrid
status of the plant. In a tetraploid, however, the chromosomes may be
homologous (have the same DNA sequences more or less) in *part* of a
chromosome, but not in the whole length. That part pairs with what it
matches in another. The remainder of the chromosome may match a
different chromosome, so those parts pair--you end up with a "trivalent"
association--three chromosomes together. The fourth may end up
unmatched, or even pair up with a part of one of the three that are
daisy-chained already. That's a "quadrivalent" or "tetravalent"
(depending on the writer) association. What further complicates this is
that not all cells that lead to pollen grains or ova necessarily behave
the same.
When any of these oddball
groupings occur, it blows the probability pattern to pieces.
The basic square is also much
larger. Instead of having a two-by-two square with four possible genetic
outcomes for any one locus, the square has four possible conditions on each
side, so the basic square has sixteen possibilities, not four.
Then, there is another factor
entering in. How the ratios work depends on how far the gene locus is
away from the centromere--the constriction between the two arms of the
chromosome, where the chromosome gets grabbed by the mechanism that draws them
into the array necessary for either meiosis or mitosis. The farther the
locus, the higher the probability of crossover occurring between the locus and
the centromere, which affects the probabilities. There's even more to
that than this suggests.
Since we don't know for sure just
how many genes are involved in SA behavior, this gets radically complicated by
number. The potential is a "cube" of 4x4x4, or 64 basic possible
outcomes if only two loci are involved.
If we succeed in pinning down one
basic fact--the number of loci being one, or more than one, we will have
accomplished much in the SAGE project.
If we succeed in generating
numbers that reveal presence or absence of dosage effects, that would be a
major step forward also.
And so on. Any solid,
number-supported, result at all is worth the work we are
contemplating.
Hang in there and keep
speculating. You've got a fertile imagination.
Neil Mogensen z 7 western
NC
