Hi All,
I've been enjoying this conversation quite a bit. I would like to
throw in some ideas and concepts from the statistical genetics
literature ( http://goo.gl/kXkTxx ), and I would like to throw out
some data and thoughts. Some of it is the same as at the Glenn
apiaries site, http://www.glenn-apiaries.com/genetics.html. The rest
is more general coming from a human genetics perspective.
Thoughts: in genetics, there are several key concepts, some of which
aren't getting play here. The first is how genetic variation works.
In genetics, we have a concept of a population of individuals (3 key
ideas), and (idea 1) there are different versions of each gene that
pop up at a certain rate (mutation rate). If these different versions
of genes don't kill the individual, they can enter the population and
be transmitted to other bees in future. Next, is something that
wasn't intuitive to me at first, but makes sense in the larger
picture, and that is (idea 2) recombination. (I'll just let you read
it http://en.wikipedia.org/wiki/Genetic_recombination) Recombination
is when chromosomes swap parts with each other, creating a chromosome
in the offspring that is a mix of the two parental chromosomes.
Finally, just because you have a particular gene (idea 3), doesn't
mean you have a certain phenotype (eg If you have the Huntington's
disease version of a gene, you're 99%+ likely to develop Huntington's;
but if you have the "bad" version of the heart attack gene (apoE3
maybe?), you're only 30-40% likely to have a heart attack; the idea is
called penetrance, http://en.wikipedia.org/wiki/Penetrance ). I think
penetrance and recombination are some of the gray parts of genetics
that people don't talk enough about.
(key idea) The population is not static - there is variation (in the
statistical sense) within every population. There is great movement
in the frequencies of versions of genes (this is measured with the
test for HWE in diploid things,
http://en.wikipedia.org/wiki/Hardy%E2%80%93Weinberg_principle ). No
matter how hard you try to control the genes in your population, there
are new combinations of genes on a chromosome at every generation.
That's part of the great joy of why we have existed for millions of
years (as environmental changes like weather and tree distributions
(due to disease) that one year's 'weak bees', could be the only ones
left in a couple generations of extreme weather.)
This is where the dissonance comes in, because if you want to
commercialize an operation, you want to reduce the variation in your
bees so that you have more desired behaviors and less undesired
behaviors. This is a mixture of genetics (think VSH) and environment
(managing the bee to comb ratio, feeding, etc). So as a commercial
beekeeper, you want less genetic variation because it reduces the
variation in your managed hives. One good extreme example of this in
population genetics is the idea of a "recombinant inbred line" (it's
more a mouse thing in practice). If you've heard of hybrid lines (I
have heard of starlines,
http://www.bushfarms.com/beesraces.htm#starline or maybe midnights?),
then you have the idea of 2 inbred lines (2 populations with very,
very low genetic variation) they come together to create individuals
with almost uniform desired characteristics. (a great business
model). But it's really difficult to maintain inbred lines (
http://link.springer.com/article/10.1007/s00114-013-1065-y )
Tarpy has a good paper (or at least reference) showing some evidence
(using microsatellites as data) that there are 3 genetic populations
of honeybees in the US: a west coast population coming from breeding
operations there, a southeast population coming from breeding there,
and "feral" bees (I think Rangel did work on feral bee genetics? It
might have been a different post doc?). I think this was before the
Russian breeder and NWC were going? Or they weren't in the results
for some reason?
I would suggest thinking about the conversations of local adaptation,
genes, etc in terms of population and not in terms of individual
hives. From what I've seen, most commercial beekeeping operations are
lucky to have a queen last more than 2 years, so that brings about the
question: where are their genetic populations? Are they buying queens
from breeders, or raising their own? With local "feral" bees coming
from populations with very high genetic variation, you would expect
some very strong hives every year. But you also have some total bombs
(dead outs) that nobody ever seems to talk about. So any conversation
about feral bees should be focused on outcomes with respect to the
population. The top ~ 5% of the yields from a population of ferals
should beat the yields from all producer bred populations, but
commercial beekeepers care about the average yield and the cost of
management, a different business than sideliners/hobbyists. (eg
managing more than 500 hives vs managing less than 50).
I have seen a couple long term sideliner operations (200-500) try to
get into queen breeding. It takes a couple years for their breeding
operations to start to change (drone flood?) the distributions of
genes in the populations in their area, but as the years go by, they
have bees very well suited to their environment and management
practices. [a related question would be: if I have n=500 hives and
want to breed my own queens, how does the number of breeder queens and
number of grafts of each affect the distribution of versions of genes?
How can I use ideas of population genetics to improve my stock? I
would suggest a book like Falconer, http://goo.gl/KMydS9 if you're
interested in that.]
Sorry for the ramble. I hope it's useful to think about the science
behind some of these observations. The thoughts apply equally to
thinking about the population genetics of varroa. (on another note,
does anybody know if there has been any use of sterile males in varroa
control? I don't know the lifecycle of male varroa or about genetic
diversity in varroa- but I had the thought of screwworm eradication in
mind, http://www.fao.org/docrep/U4220T/u4220T0a.htm ).
Best, Blair
***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software. For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html
|
