Using segregation to
recover traits and create new varieties
Let's now look
at how one can use that segregation from the previous
page to select and create new true breeding
lines from any hybrid.
The F1 hybrid is self
pollinated and creates a segregating F2
generation.
X
|
ovule
|
ovule
|
pollen
|
heterozygous
|
homozygous
dominate
|
pollen
|
homozygous
recessive
|
heterozygous
|
| phenotypic
ratio |
3:1 |
of
4 possibilities, 3 regular leaf
plants
and 1 potato leaf plant |
One in every
four plants should be a recessive potato leaf. If all one
wanted was a potato leaf plant, one would just select for the plant(s)
that
show the trait and that would be it . Recessive traits are easy to
recover
because as soon as you see that trait you know it is "fixed" because
the
only outcome it can have in the next generation (assuming self
pollinating only) is another recessive. Like as stated on the
first page:
Once a recessive trait is recovered, that
recessive gene will be in all of that individual's offspring.
Fixing or recovering a dominate trait in a population is not as easy.
This is because
when one sees the dominate trait in the F2 (and for the next few
generations) one can't be sure if the genotype is heterozygous or
homozygous.
To recover a dominate trait it will require one to select for it and
grow out till at least to the F7/F8 generation for the reasons
illustrated on the page about
segregation. This will process rids the population of ambigous
heterozygotes so that the line will eventually
breed true (by becoming homozygous).
Though it may be
easy to recover a recessive quickly, one must remember
that it isn't just that one recessive gene segregating but many others
which may or
may not be desirable (like plant size, foliage cover, yield, etc..).
For this
reason one should continue to carry out that line with the desired
recessive trait (through continuous
selfing each generation) to the F8 to ensure that all other gene pairs
are fixed by removing most, if not all, of the heterozygotes.
Using Segregation
Now let's
consider how one can use predictable segregation patterns to help
purposely "create" a new
line.
Because one can
estimate what the potential outcome will be in a given F2 population,
one can also
predict how
many plants one needs to find a given number of recessive traits.
Lets say one has
two plants that each have separate characteristics and would like
to combine both of those characteristics into a new line. One plant is
a potato
leafed variety with red fruit. The other plant is a regular leaf
variety with yellow fruit. How many plants do we need to find one plant
with both traits - a potato leafed with yellow fruit?
Potato leafed is recessive and from other lists
know that yellow fruit is recessive to red fruit.
So:
- Two traits
- both
recessive
- yellow
fruit and potato leafed
Remember that in
an F2 one can expect that 1 in every 4 plants will be recessive.
If we are trying to recover 2 recessive traits in the same plant, there
is a 1 in 4 chance for each trait. A 1 in 4 chance for a potato leaf
type and a 1 in 4 chance for a yellow leaf (4 and 4). It will take16
plants to find one
plant with both traits.
From
the 4 chances for each gene 16 types are found.
Only one plant (bottom right corner) is both yellow fruited and potato
leafed.
So to get just
one plant with both one needs to grow at least 16 plants. However, one
may
want to select more than just one plant with both traits. So, by
growing out
32 plants, one should find at
least two plants (potato leaf and yellow fruit). Then one could select
for other traits - like plant
habit,
foliage cover, yield, taste - among those two lines and have more
opportunities to select from the best
of those other segregating traits till stable.
What if one wants to look at more than just 2 genes?
#
of recessive traits
desired
|
Chances
for each
|
#
of plants needed
to find one (with all)
|
1
|
4
|
4
|
2
|
4
x 4
|
16
|
3
|
4
x 4 x 4
|
64
|
If one is looking to recover even more traits simply use
the calculator below to determine how many plants will be needed to
find one individual with the number of traits entered.
Calculate Required
Number of Plants Needed to Recover n Recessive Genes
Use the "+"
and "-" buttons to toggle the number of genes
Notes:
- one
will see the most segregation in the F2 and F3 lines. These are the
best generations to try to recover traits desired. Some of it will be
serendipity but by applying the know potential outcomes and selecting
the desired traits early one can easily develop a new line
- always
take careful notes each year about what was observed in the various
lines. Since this will occur over a matter of years pictures can help
as well.
De-Hybridizing a Hybrid to make a
stable true-breeding line.
Let's say you want to take a hybrid and make a reasonably close form of
it so that one can save seeds of it and have it come "true" year to
year. The process is really quite simply. Just use the segregation
method and at the F2 begin
selecting for the plants that best represent the
original parent. Continue to do this for 6 or 7 generations and
one
will have a line that will be stable and true-breeding and somewhat like the hybrid parent. I
say "somewhat" because it will be impossible to retain any heterozygous
traits expressed in the hybrid that may have had made it unique and
with hundreds of potential gene combinations, it is nearly impossible
to grow
enough plants to best represent the original hybrid (just try running
that gene calculator up to the number "50" and see how many plants it's
going to take just to get one that has all 50).
Other Information
These pages only discussed simple inherited dominance. There are other
several other forms of inheritance I have not covered (such as
incomplete dominance and co-dominance). This technique mentioned above
is just one form of cultivar/variety development that lends itself well
to working with an inbreeder like tomato. See these for more
information about other kinds of inheritance or breeding strategies.
Resources for determining inheritance of genes:
