Selection for se in corn - selfing vs mass populat

[spero]

Carrying on from the triploid discussion -
The central question here is which method - selfing or selecting in populations - will better get us to the pure se condition in an open pollinated corn.
In creating a viable op, we need as much diversity as we can maintain to keep the population productive. This tends to require large populations.
In creating an inbred line, we care more about purity than vigor, so that tends to favor inbreeding methods.
I question 16 or even 50 plants. The CW I have heard is 250. One reason hybrids have such gain may be that the op populations havn't been big enough. -JS

[spero]

The question is, can we get to pure se without inbreeding ?
Inbreeding is a lot of work. Individual seed selection is a lot of work.

Is it correct that half the plants will be 1/2 se throughout, the other half random by pollen, while the other half of the plants will be 1/2 su throughout ?

How many plants doth a viable (or optimal) population require ?

[DarJones]

There are valid concerns about inbreeding with corn. Vigor goes down the potty fast as do some other significant traits.

The minimum population to keep a diverse breeding line of corn going is about 8 plants. But this presumes that you have a diverse breeding line to start with. Lets say that your friend Joe gave you 500 seed of an old OP corn that has been in his family for years and presume that he told you it has always been grown in isolation so it could be kept pure. The diversity in those 500 seed could be less than you would find in just 3 seed of a highly diverse line. This is the reason you so often hear that corn requires a population of at least 250 plants to maintain the line.

I am NOT advocating using only 8 plants to maintain a corn line folks, this is just a minimum number. If you want to maintain a line of corn long term, you should know just how much diversity it contains and plant accordingly. The less diversity to start with, the more plants you need just to maintain what you have.

DarJones

[silverseeds]

Well if you do pull off this Se thing... perhaps, once your showing it will work, find someone interested who will make a separate line. then when you work those two together later, you should bring the vigor back Id imagine......

[Joseph Lofthouse]

The su and se genes are on separate chromosomes...

So they are inherited separately and not as a unit, but the
math is the same. (If an indian corn and a se+ sweet corn
are crossed only 6% (25% of 25%) of the kernels will contain
su/su and se/se.)

It is easy to select for sweet corn by sight due to the wrinkly
kernels. It is less easy to identify se kernels by sight, especially
since the pericarp is from the mother plant.

I want to believe that kernels expressing the se gene are more
rounded, and more finely wrinkled, and less angular than regular
sweet corn seeds, but whether or not it works out that way in the
real world remains to be seen. (I have received seeds from several
se breeding projects and have sorted them according to this theory.
I'll measure the sugar in them this summer and see how accurate
my guess is.)

In theory, if there are no phenotypic differences for se+
it is impossible to eliminate the anti-se gene from a
population only by selection because the most you can hope for
is to half the concentration of the anti-se gene with each generation,
(if you do perfect selection every generation).

To get a pure se+ line we have to provide a pollen source that is
pure se and that does not have any anti-se pollen in it.

I see two ways to do that...

1- selfing

2- back-crossing.

The back-cross only has to be a pure se+ source, it does not have to be
any particular variety.

That is simple to say... In practice there are a lot of troubles....

Do you trust the seed company to properly label your seed stock? What if they
call something se+ and it is only se, or su, or sh? What if two seed companies have the same
name for different varieties? What if the clerk mixed up the labels on the seed bins?
Or spilled seeds on the floor and put them back into the wrong bin?
To do a back-cross right you'd want to get a large lot of seed from a trusted reliable
source, grow it out and test the children the next year to make sure it really was se+.
And store the seed to maintain it's viability for years.

So lets assume we have selfed or back-crossed a plant.
How do we know it is
homozygous for se+? The only way I can think to figure it out is to plant
a couple dozen of the seeds from each offspring
cob and measure the sugar in them. If every cob in the child population
is high sugar then the cob was homozygous for se, but if some of the
children are high sugar and some low sugar and some medium sugar
then we know that cob was not homozygous....

Whether I use selfing or back-crossing I wouldn't do it on the F1 population.
I'd wait until after the F2 generation has segregated and I was able to select
the cobs with the highest sugar content (which are likely to come from se+
mothers contaminated by some anti-se pollen.).
This greatly minimizes the amount of selfing/testing that needs to be done.

I am intending to do a back-cross this summer on Astronomy Domini, Sparkler,
Fiesta, Ashworth, etc... just in case any of them happen to be se+

But if there are identifiable phenotypic differnces for se+ kernels,
then no back-cross or selfing would be necessary.

Selfing or backcrossing would require immaculate record keeping.
Selecting by phenotype would not require any record keeping.

Regards,
Joseph

[Joseph Lofthouse]

Or I could use a good enough strategy...

For seed that I grow for my own use on my own farm,
as long as I only plant seed from cobs with the highest
sugar then there will not be any duds (regular sweet corn),
and the population will get more sugary each year
as 1/2 of the anti-se gene is eliminated each generation.

If I sold seed, then my end users are no worse off than if
they were buying hybrid seeds from THE CORPORATION.

This doesn't meet the promise/allure of open pollinated
seeds, but in the short term it might be an acceptable until
the back-cross/selfing/phenotypic-selection was completed
and verified by testing.

Regards,
Joseph

[Joseph Lofthouse]

Mar 9, 2010 23:01:05 GMT -5 Joseph Lofthouse said:

To do a back-cross right you'd want to get a large lot of seed from a trusted reliable
source, grow it out and test the children the next year to make sure it really was se+.

Or maintain my own population of se+ seed and never add any new genetics
to it until after it has passed a selfing/segregation test.

Regards,
Joseph

[spero]

If we are seeking an open pollinated, vigorous se variety, we need to select for homozy in the se gene, while maintaining heterozy in as nearly all other locations as possible. This argues against both selfing and backcrossing, both of which will make the population less diverse in many locations.

If we can eliminate the half the plants that have an su, and do nothing else, we gain generation by generation as Joseph said. If we can effectively select kernels at drying stage, and know that the kernels on a given cob are half or all se, I think we can get to sese.

What I had been doing was selecting the slowest drying 10 - 25 kernels from each cob, therefore only gaining some in se. But if one of the brix test methods can eliminate half the ears, we should get there fairly quickly.

From Dr. Jack Juvic -Plant Geneticist U of Illinois:
"slower drying is a fairly effective method to select for homozygous (se/se - not heterozygous Se/se - with only one dose of the recessive gene) se kernels. Since it is recessive you need to have 2 doses ..."

If we are creating inbreds to later cross out to make f1's, none of this applies.

[DarJones]

I dug out quite a few of the answers re breeding sweet corn and also dug out the name of someone who knows all you can ask about sweet corn breeding.

Bill Tracy university of Wisconsin.

agronomy.wisc.edu/index.php?c=2&p=1&facid=66

I will try to contact him later today to ask a few questions.

DarJones