breeding OP's - final generations advice please

[bower]

So after reviewing my results this year, I have seen a lot of segregation in F2 and again in F3, and would like some advice about the generations 4-7, what kind of segregation to expect and how many plants should be grown to capture the traits desired, which I selected in F3.

My impression is that this may be different depending on the cross. In some the overall traits appeared fairly stable even in F3. I have also recessive/desired traits selected in F3, but while leaf type and fruit color are expected to remain, there are many traits which are additive and still expected to vary. Some crosses appeared to have fairly stable, desired traits in F3 and easily selected for the best from 4-6 plants.
I also know linkage can play havoc, and I'm seeing that I think with growth habit segregations in F4. But I would like to know how many plants you would grow in gens 4 - 7, for segregation in:

fruit quality = taste and texture
fruit size
earliness
productivity
other traits that pertain to plant health? I have not done any selection on the basis of health except this year putting some plants outdoors there were perhaps differences.

What are your thoughts on stability, perceived or actual?

Every advice or comment about your experience would be much appreciated!

[carolyn137]

Originally Posted by bower

So after reviewing my results this year, I have seen a lot of segregation in F2 and again in F3, and would like some advice about the generations 4-7, what kind of segregation to expect and how many plants should be grown to capture the traits desired, which I selected in F3.

My impression is that this may be different depending on the cross. In some the overall traits appeared fairly stable even in F3. I have also recessive/desired traits selected in F3, but while leaf type and fruit color are expected to remain, there are many traits which are additive and still expected to vary. Some crosses appeared to have fairly stable, desired traits in F3 and easily selected for the best from 4-6 plants.
I also know linkage can play havoc, and I'm seeing that I think with growth habit segregations in F4. But I would like to know how many plants you would grow in gens 4 - 7, for segregation in:

fruit quality = taste and texture
fruit size
earliness
productivity
other traits that pertain to plant health? I have not done any selection on the basis of health except this year putting some plants outdoors there were perhaps differences.

What are your thoughts on stability, perceived or actual?

Every advice or comment about your experience would be much appreciated!

First, what varieties are you crossing, viz wide or narrow crosses for it makes a difference as to how long it will take to stabilize something in terms of genetic segregation.?

And I think if you go to Keith Mueller's superb website you'll find most of the information you need as to most of your questions, especially about segregation.

http://www.kdcomm.net/~tomato/

I've never bred for OP's but have dehybridized some F1 hybrids so needed the same kind of information and Keith addressed that very well.

Almost forgot,but when you said breed for OP's, does that mean you are introducing various wanted genes,one at a time? I ask b/c that makes it much more difficult since you have to assess that incorporation at each step, and that's difficult since it means challenge experiments at every step. Which are done by registered labs who do that and that's big money.

Carolyn

[bower]

Hi Carolyn, thanks for your reply.

The crosses I've been working on are described in my posts, most recently Hilberts Grand Tomato Hotel ( the place where there is always room for more guests, even if it's full )
I understand what you're saying re wide vs narrow crosses, some of the crosses described are not close to the goal but a few are actually pretty close to what was aimed for.
Also some are at F3 and not at all stable, one is at F5 but not as stable as I hoped... etc.
It is a baffling process and full of surprises, so I'm hoping for some general insights about the stability of the additive or QTL traits, and how to reckon how many plants to grow when there isn't a simple recessive/dominant to select for.

[KarenO]

Hi Bower,
All I know is my own way and that is to grow as many as I have room for
It's hard to resist the temptation to bite off more than I can chew and work on more crosses before I finish my current ones. I only save seed from one plant of each growout beyond F2. When you find differences F3 and up which happens frequently and you save seeds from those plants they will act like another F2. That is, if you save Seed from an F3 plant that is different than the f2, it is like saving another F2 IMO. If i were to Keep doing that generation after generation, saving seed from plants that are slightly different, it will never be stable.
The most important thing is to do a large f2 growout, select one you love of that group and grow out seed from that one looking for F3 that are the same. Choose only one of the F3,grow that out and so on.
If there is not a selection in your F3 that is the same as the F2 you loved then best to grow out another batch of F3 seeds. To me, the thing you need is a GREAT F2 selection that has everything you are looking for. If it isn't great at F2, it won't become great with further selection in my opinion. In my projects so far, the key is in the F2 selection. Simplistic I suppose but it has worked for me more than once. It is easier with recessive traits for sure.
It's easier to go back to the drawing board and make a different cross or to do another F2 growout looking for a better segregant to stabilize than to go forward with anything you are not completely happy with hoping it will get better. It might but IMO every additional segregation sets the stabilization process back a full generation.
Example: my F3 bicolour PL heart threw a green and pink bicolour heart segregant. . I saved seeds of course and planted them this year but I consider it A new segregant and it needs additional time to stabilize it.
Highly anecdotal and unscientific perhaps but it's my way. Have a specific goal in mind, make the cross, look for your goals to be met in the F2 and go forward selecting only for those traits. If you find additional segregants beyond F2 that you want, save seed and grow but they are new segregant and will take more time.
Stabilize must mean selecting for maximum sameness generation after generation so the decision of what is wanted needs to happen right away.
KarenO

[bower]

Thanks KarenO: your POV is appreciated.

What I'm wondering is how much segregation you continue to see in F4, 5, 6... after you have selected the plant you love from F2 and F3 generation. How many plants did you need to grow to find one identical to your chosen prior generation? Or expressed as a ratio, how many of four or six or ten plants are so similar by F4, 5, 6... that you can't really tell them apart?

As Carolyn pointed out, that would depend to some extent on the cross, how wide ie how much variation is there that has to be rendered homozygous. Obviously if the parents are very similar then there will be more stability early on. Also if the parents are not themselves stable, there's going to be more variation, potentially, for generations to come.
And chance plays a part as well, in the segregation. So for example, I have two sibling lines from a cross between an F3 and an F1 - one of them had a high degree of similarity to the selected F1 (skipper) - meaning that 4/6 F2 plants had the same fruit shape and size, and similar quality. One had the recessive pink-black trait but otherwise entirely similar. But the other sibling F2 was all over the map, none of them matching the shape and size of the original selection (rodney). So it stands to reason, I need to grow more of that F2 to find the one I was looking for, and I expect it to take longer to stabilize because that line contains a lot of divergent traits.

[KarenO]

My unscientific approach is simpler. with the heart crosses that seem to be stabilizing quickly, I used two known stable varieties and crossed them. Picked only 3 of the best F2's and brought them forward selecting as close to identical offspring from a F3 grow-out of only 6 plants each. I chose only 3 because that was all I figured I could deal with in the space I have. I did keep seed from a lot of the other F2's because they were all quite good, a few unbelievably good beefsteaks but I chose to only do the hearts because they are my favourite and there are so few PL hearts.
In my PL cherry project, I thought I was crossing 2 stable varieties again. Wrong. Turns out the cherry I crossed was unstable and the progeny continue to toss off oddities in F4 but that is from an oddity that turned up in the F3... so really it was still an F3 according to my simplified logic... meanwhile the 2 origional selections, one a dark green shouldered purple and the bright pink cherry I origionally selected are looking good going into F4 and only a small number of plants of each have been planted. By F3, according to the probabilities 75% should be looking the same but that is only if both parents are perfectly stable... Add in 2 unstable parents and the number of plants or even the total numbers of grow-outs I'm not sure can even be calculated for certain.

I hope the real genetics wizards can help you, clearly that I am not and my experience is limited to my own little projects but these are my thoughts and observations. I am working with a lot of recessive characteristics and so it is easier to see the differences and once the recessive trait is expressed it is fixed which also makes it easier.
Karen

[bower]

Thanks, it is helpful to hear what your experience has been.
Some of my crosses are straightforward as well with stable parents, and I did see the same thing, some recessives turned up in F3 but overall there was a pretty high degree of similarity in fruit quality, size and shape etc. 75% would be a good ballpark.

I was pretty surprised to see that 75% ratio go down to 50% in my Black Nipper F4, though, and one of the reasons I wanted to ask this question. Size and taste quality seem to be segregating with the growth habit - smaller fruit, and less tasty on the smaller determinates... maybe effects of linkage there. Only four plants, so maybe the ratio was affected by chance too.