I can live with those odds!

Share your breeding experiments and crosses you're working on
Post Reply
User avatar
Pippin
Reactions:
Posts: 149
Joined: Fri Mar 20, 2020 6:49 pm
Location: Finland

I can live with those odds!

#1

Post: # 34026Unread post Pippin
Sun Nov 15, 2020 11:50 am

I am surprised how good odds one has of keeping a recessive trait when selecting a dominant phenotype from a progeny.

Let’s assume one values the first recessive trait very high, say a compact growth habit. The second recessive trait, say tangerine fruit colour, is just not there. Not at least in the perfect compact plant.

Now one selects compact plants with red fruits for next sowing, here are the odds for still getting tangerine in next generation: 1-1/3 = 67%.

But if one selects two red fruited compact plants for next sowings, one increases the odds: 1-(1/3)x(1/3) = 89%.

ChoosIng three Increases the probability to 96%.

The same principle applies to any generation which seggregates the wanted recessive trait in siblings that you don’t want to choose for some reason. It does not need to be F2 generation, for example.

It may seem strange maths at first until you realize that there are only three options: TT, Tt, tT. Red fruited plant cannot be tt.

Do you agree with these odds or did I make some mistake? It is long time since I was in school. :D
BR,
Pippin

User avatar
bower
Reactions:
Posts: 5474
Joined: Thu Dec 12, 2019 12:44 pm
Location: Newfoundland, Canada

Re: I can live with those odds!

#2

Post: # 34030Unread post bower
Sun Nov 15, 2020 2:04 pm

You're right that the chance of your selected red fruited F2 having the recessive is 2/3, since we know the fourth possibility didn't occur (tt).
I'm not so sure about the math for what follows though.
I believe you increase your odds of getting a Tt plant when you select more than one F2 plant to save seeds from.
However for the F3 generation from any given plant, you're still with the same math for getting a tt from a Tt, and the chance of that is 1 in 4 but only if the plant is a Tt and not a TT. So you're actually multiplying the 67% chance by the 25% chance in terms of your probability, which is naturally a bit lower than you would have, if you certainly had a Tt plant to work with.
The bare reality is, you will have to grow at least 4 plants to find out if you have the recessive. I usually grow 6 to make reasonably sure I will find it if it's there. (Someone at TV did a really nice math that shows how the element of chance weighs in, and you're definitely safer with a few more than 4 to be certain.)

If you saved more than one F2 plant that might be Tt, then your odds of having a Tt will go up but you still have to grow more plants, at least 4 per individual that you saved. None of them are better than Tt, therefore your math is always based on that, with the caveat you might have TT and if so will not find the recessive because it isn't there.
Still the 67% is quite good for finding a second recessive in the next generation, I have done that a few times and even had less likely recessives turn up and surprise me. So don't be daunted! :)
AgCan Zone 5a/USDA zone 4
temperate marine climate
yearly precip 61 inches/1550 mm

User avatar
wykvlvr
Reactions:
Posts: 460
Joined: Fri Jun 26, 2020 11:36 am
Location: Southeast Wyoming

Re: I can live with those odds!

#3

Post: # 34032Unread post wykvlvr
Sun Nov 15, 2020 2:17 pm

Actually there are only two options TT or Tt. Tt and tT are the same thing when it comes to segregating for the next generation. IF you get a plant that is Tt you hit the jackpot as you could get compact and tangerine but most would be compact and red in the traditional 3:1 ratio. On the other hand if you end up with one or more that are TT you are SOL. The problem of course is that you have no visual way of telling which is Tt and which are TT. AND of course you still have to grow out LOTS of the next generation to determine if your selected plants are TT or Tt. AND your next generation would present the same problem. Which are TT and which are Tt? Of course once you get a single plant that is compact and tangerine you know that they carry all the recessives you want.... then you can start to select for taste, etc.

Folks tend to forget that ratios etc are for HUNDREDS of plants or whatever you are breeding. For example Reg leaf is dominant over carrot leaf, reg leaf is dominant over potato leaf, carrot leaf and potato leaf are on separate locations so carrot x potato = all regular leaf in the F1. F2 can segregate out and produce all three leaf types... In theory that should be 9 reg:3 carrot:3 potato:1 that is both potato and carrot leaf ie the 9:3:3:1 ratio. The plants I have right now are an F2. First 36 seedlings were 2 carrot leaf, 1 possible potato leaf, 33 regular leaf. The next 36 are 2 possible potato, 6 regular leaf, 28 carrot leaf.... so 39 regular: 30 carrot:3 potato which is not at all the expected ratios but my sample size is too small for the theoretical ratio to show up.
Wyoming
Zone 5
Elevation : 6,063 ft
Climate : semi-arid
Avg annual rainfall = 16 inches

User avatar
Pippin
Reactions:
Posts: 149
Joined: Fri Mar 20, 2020 6:49 pm
Location: Finland

Re: I can live with those odds!

#4

Post: # 34058Unread post Pippin
Mon Nov 16, 2020 8:58 am

[mention]Bower[/mention] This is exactly what I meant. Choosing more than one red increases the odds for getting one Tt. Growing 6-10 of each in next generation was the plan in my mind. But what surprised me was that the odds of getting even one Tt is not 50% but 67%. The rest of the math is the probability of choosing two or three TT’s and taking a negation (e.g. NOT(TT AND TT)).

[mention]wykvlvr[/mention] I was trying to demonstrate with the Tt/tT syntax whether the genes come: pollen or egg. Sure, the phenotypes are the same.

BTW, I made a test sowing early this year but did not observe the 1:3 ratio on a phenotype that should be controlled by a single recessive gene. I started thinking if I made a systemic error in my way of choosing the seeds. Then I realize that I tend to choose healthier looking larger seeds when sowing. I did not think that the size of seed could correlate with the property I was looking for. I needed to force myself to also sow smaller seeds and grow out the weakest looking plants in the actual main sowing. Later this year I found a scientific article saying that this is possible: seems that some compact growth habbit genes may express themselves as smaller seed than their normal siblings. Also, the heterozygote plants seem to be shorter than the normal growth habbit parent, kind of intermediate phenotype. I suppose the dominance is incomplete.
BR,
Pippin

OhioGardener
Reactions:
Posts: 279
Joined: Mon Dec 23, 2019 10:58 am
Location: SW Ohio, Zone 6

Re: I can live with those odds!

#5

Post: # 34064Unread post OhioGardener
Mon Nov 16, 2020 10:28 am

Interesting topic. It sounds similar to what I'm working on. This year I started my first tomato breeding project by crossing an indeterminate tangerine to a red dwarf and will be growing out what I hope will be F1 plants in 2021. It took some trial and error to get the hang of pollen collection and emasculating flowers at the right time, so I'm hoping my two tagged fruits that set will be actual crosses. I'm a Biologist and former dog breeder/exhibitor, so I have a solid grasp of genetics and selection, but this is my first time working with tomatoes. Several folks were kind enough in another forum to point me to a spreadsheet of known alleles. Tangerine is a simple dominant/recessive trait, but it looks like there may be quite a few different mutations/alleles that can affect plant size which may complicate probabilities if one doesn't know which dwarfing gene or genes they're starting with. I was also initially hoping to be able to harvest F1 seeds early enough to squeeze in a large planting of F2 seedlings and be able to visually cull for tt (and hopefully dwarf) at the flower stage. However, I see that there are other genes described as having an effect on the intensity of flower color. Along this line, I noticed that some of the existing tt varieties I grew out this year didn't have anther colors that were visually distinct from some of the red varieties I was growing, so this might not be practical. It'll be a fun learning experience for sure.
Debbie

User avatar
wykvlvr
Reactions:
Posts: 460
Joined: Fri Jun 26, 2020 11:36 am
Location: Southeast Wyoming

Re: I can live with those odds!

#6

Post: # 34065Unread post wykvlvr
Mon Nov 16, 2020 10:35 am

DUH should have realized that about the tT and Tt, been spending too much time in the eat you alive genetics group on Facebook ;-)

I do have an interesting little plant in my collection of seedlings, it is very short and compact but sadly I didn't pay much attention to seed size when planting. I do know that there was only about half a fingers width between the cotyledon fake leaves and the first true leaves, there was about the same distance between his first and second true leaves... it is an incredibly short, bushy compact plant. In other words Tiny has the shortest internodes I have seen on any of my plants and that distance could be a visual way of selecting for compact plants...

This photo is from September 5th
Image

172X F2 seedling E2

This is a current photo of the same plant notice how compact it has remained...
Image
E2 aka Tiny by spindledreams, on Flickr
Wyoming
Zone 5
Elevation : 6,063 ft
Climate : semi-arid
Avg annual rainfall = 16 inches

User avatar
Pippin
Reactions:
Posts: 149
Joined: Fri Mar 20, 2020 6:49 pm
Location: Finland

Re: I can live with those odds!

#7

Post: # 34110Unread post Pippin
Tue Nov 17, 2020 12:19 am

That is seriously compact tomato. Never seen such a thing. Looks very healthy little plant.

I tried finding the reference on this seed size issue but it seems that I did not bookmark it. Anyhow, I am not claiming I have this trait. It only made me more aware of the selection I may be doing without being aware of it.

I had my first F2 generation this summer and I am just in the beginning of my tomato breeding journey. I read from another forum that people tend to select F2’s that are different from F1 (with all dominant traits), like fruit colours other than red. I started to think that the fruit colour is probably the easiest thing to fix, so I tried very hard not to overlook red fruited plants. I still think my math is correct or at least demonstrates an important topic.
BR,
Pippin

User avatar
wykvlvr
Reactions:
Posts: 460
Joined: Fri Jun 26, 2020 11:36 am
Location: Southeast Wyoming

Re: I can live with those odds!

#8

Post: # 34130Unread post wykvlvr
Tue Nov 17, 2020 1:42 pm

Grin he is amazing, just as healthy as my larger plants in that set of seedlings and is the only one like him out of 64 seedlings. Alas he has not flowered, not even produced flower buds yet so will be a very late tomato when he does fruit. The other plants I kept that are the same age as him have tomatoes or at least flowers. I have 4 plants who should have ripe tomatoes in a couple of weeks (high on the keep list) 4 that are just starting tomatoes or at least blooming (kept due to leaf type or plant size) and then there is Tiny....

I do have to admit the seedlings I am raising are not my breeding. They were sent to me as part of the Micro Tomato Diversity Project so I have no idea what is behind him. But I am having fun.

Yep your math even if slightly off does demonstrate that it is not as difficult to recover recessive traits as many folks think. AND you could get lucky in the F2 and get a plant with both features you want.
Wyoming
Zone 5
Elevation : 6,063 ft
Climate : semi-arid
Avg annual rainfall = 16 inches

Post Reply

Return to “Tomato Crossing/Breeding”