Micro dwarf x Indeterminate cross ideas.

[KaguyaCloud]

Very interesting on your observations on vigor and fruit set. I actually have the opposite problem with micros when they are grown in small pots! They almost always become top heavy with the fruits they set and fall off onto their sides a lot. Since they're stems are much shorter and their trusses are more compact, they can support quite a lot of fruit with any breakage. It also helps that the stems are quite thick as well.

In terms of flower morphology, I think the micros I am currently breeding inherited the less delicate pistils from the indeterminate cherry plant. They are much easier to de-masculate than normal micros. However, it seems consistent that their anther cones is about half the length of a normal tomato flower anther cone.

Parthenocarpy is actually quite easy for me to see since my micros should have clear skin, yellow flesh, and green gel. Fruits with viable seeds form a green gel coat around them. All I need to do is hold the fruit over a bright light and see if there's any green.

[KaguyaCloud]

Eliminating potential micro gene locations for micro-dwarf characteristics based off of personal observations of my Joro breeding project.

Non-micro related traits:
Chromosome 1: Fused sepals(SlAGL6). Clear skin(y).
Chromosome 3: Yellow flesh(r). Micro-Tom and Micro-Tina exhibit the wild type red flesh.
Chromosome 7: Lateral suppressor(LR). F2 and F3 inherited extreme axillary shoot generation from parent Rosella Cherry while maintaining compact size. This chromosome definitely does not have the early stem termination gene. https://www.esalq.usp.br/tomato/ls.pdf
Chromosome 8: Green flesh(gf). It is more that the gel color remained green instead of the flesh.
Chromosome 10: Uniform ripening(U). Green shoulders did appear, then disappear as the tomato ripens.

Definite chromosomes with micro traits:
Chromosome 2: Dwarf(d)
Chromosome 6: Self Pruning(sp)

Traits that have not been bred out from the micro parent Jochalos:
Chromosome 2: Multi-locule fruit. 3+ locules. Already present in micro parent Jochalos. This likely will not be bred out due to being within the same region of the dwarf gene.

Chromosomes that yet to have traits identified:
Chromosome 4, 5, 9, 11, 12.

Traits that I could not find chromosomes to that have been bred with the micro phenotype:
Branching inflorescence.
Long sepals

[KaguyaCloud]

I did some calculations on probabilities of hybridization after growing multiple generation lines. Initially I believed that hybrid crosses or F1 generation plants had evenly shared traits between both parents. But due to the complexities such as dominant/recessive genes, incomplete dominance, codominance, epistasis, crossing over etc, it isn't really a clean 50/50 split on traits being expressed.

However, the endpoint of creating open pollinated tomatoes is that all the chromosomes end up being homozygous. But, what I think matters more is what exactly is being homozygous. For example, assume that we want the flavor traits of Rosella Cherry and the micro growth habit of Jochalos, what is the
average probable distribution of parental homozygous traits within each generation after the F1?

Het. = Heterozygous
R = Homozygous for Rosella Cherry parent
J = Homozygous for Jochalos parent

F1: 100% het.
F2: 25%R, 25%J, 50%het.
F3: 37.5%R, 37.5%J, 25%het.
F4: 43.75%R, 43.75%J, 12.5%het.

As you can see, at the F2 generation, only about 25% of the flavor genes have full expression from Rosella Cherry in the F2 are expressed. And there are diminishing returns of obtaining homozygous traits after the F3 generation. There is a high likelihood of having 9 homozygous pairs and 3 heterozygous pairs of chromosomes at the F3 generation. In terms of probability, you will be likely to get 1 additional homozygous pair every generation after F3. That's a lot of work to only get 50% of the potential full flavor/sweetness compounds from the parent plant.

And this is where the beauty of backcrossing lies. It turns all homozygous pairs for the less-flavored parent into heterozygous pairs. It also has a 50% chance of turning heterozygous pairs of chromosomes into homozygous chromosomes for Rosella Cherry. With the sacrifice of just one generation to reselect the dwarf traits, there is a major increase in homozygous segregation for the desirable parent. In addition, the backcrossed generation starts at 50% homozygous pairs of chromosomes of the parent you backcross it with.

The math is as follows:
F3: 37.5%R, 37.5%J, 25%het.
BC1: 50%R, 50%het(37.5%R, 37.5%het., 12.5%R, 12.5%het.)
BC1F2: 62.5%R, 12.5%J, 25%het(50%R, 12.5%R, 12.5%J, 25%het.)
BC1F3: 68.75%R, 18.75%J, 12.5%het(62.5%R, 12.5%J, 6.25%R, 6.25%J, 12.5%het)

As the generations go on, it will trend towards an average likelihood of 75% homozygous chromosomes for Rosella Cherry and 25% homozygous chromosomes for Jochalos. That being said though, it is relatively simple to select towards the Jochalos traits. I am assuming that it is required that we select 3-4 chromosomes for micro traits, so it is more of a 67% homozygous for Rosella Cherry and 33% homozygous for Jochalos.

[KaguyaCloud]

Update 11/28/24


Both are F3-1 and F3-2 are flowering and successfully setting quite a lot of fruit. Flowers per truss ranging between 8-11. Growth habit for both overall is consistent with the pale F2 parent, but with shorter internodes. Green shoulders are still present in the tops of the fruit, but more prominent in F3-1.

Feeding schedule is quite heavy starting now. I am needing to feed the plants the diluted hydroponic fertilizer every 2 days to prevent nutrient deficiency. This isn't that surprising considering that there are two main stems with 50+ flowers per plant.

I have also backcrossed Joro F3-2 with the Rosella parent to see if I can introduce more flavor genes into the compact growth habit if the flavor haven't been conserved well. The potential color combinations of the future backcross will be much less crazy since the yellow skin gene has been bred out. Ideally the plan is to get every single fruiting characteristic from Rosella and to maybe increase the low sweetness levels.

[Doffer]

Most of the F1 hybrids I have created taste less good than the best parent. This would indicate mainly recessive genes.
Koralik has a dominant gene that increases the Brix level.
On the other hand, I also have the experience that the F2 generation of commercially good-tasting varieties shows considerable variation.
For example, it is very difficult to stabilize the superior taste of Sungold in a stable variety. Similarly, with Nebula, the F1 is incredibly flavorful, but in the F2, I only found about 10% super-sweet tomatoes with a Brix of 15.
This would suggest the presence of multiple QTLs that together are heterozygous for the good taste.

[KaguyaCloud]

I agree. In my situation, both have seemed to have happened. The F1 I have made had a very normal flavor, but was sweeter than both parents. It's likely the genes in my cross involving higher brix could have been additive as heterozygous. However, most of the flavor seems lacking. The intense fruity aroma with umami after taste from Rosella Cherry was extremely diluted, which likely hints to recessive traits as you stated.

Since the parent cherry variety I am using has very good flavor, I think I do not see any harm performing a few backcrosses as long as I can isolate the micro traits.

[bower]

Dominant sweetness genes makes sense to me, that's what I've noticed.
I've never tried to replicate the flavor of a parent exactly though, so I can't imagine how hard that is.
Sungold is also quite tangy besides being sweet and flavorful. Any thoughts @Doffer on dominance or segregation of the tangy note?

[KaguyaCloud]

I've never tried to replicate the flavor of a parent exactly though, so I can't imagine how hard that is.

Statistically speaking, it would only take 1 initial cross and 3 successive backcrosses to make at least 11 of the 12 chromosomes homozygous with the parent. In principle it should be relatively simple depending on what traits are being retained.

[bower]

I've never tried to replicate the flavor of a parent exactly though, so I can't imagine how hard that is.

Statistically speaking, it would only take 1 initial cross and 3 successive backcrosses to make at least 11 of the 12 chromosomes homozygous with the parent. In principle it should be relatively simple depending on what traits are being retained.

This might explain many of the variants on favorite heirlooms such as Brandywine.
Obviously doesn't work with an F1 such as Sungold.

OTOH, to be realistic about statistics and breeding, getting the segregation you wanted is far from guaranteed. Linkage causes genes to segregate together as a package sometimes, and other times not. That doesn't mean it isn't worth trying, based on statistical odds. But it might take more work than the odds suggested.

[KaguyaCloud]

I wonder how many good-tasting variations of a stabilized Sungold were thrown out just because it was red instead of orange.

Unless there are obvious phenotypes that could be distinguished early on, it definitely is a gamble in chance. This is especially so in terms of flavor(although I've heard from both Craig and Ellie that tomatoes green gel or green flesh gives excellent flavor). Currently I've been trying to see if there's any connection between certain physical phenotypes(clear skin, fused sepals, etc) and flavor. I would need to grow a ton more to see if there's any kind of segregation, however.

[bower]

There might actually be some connection between clear skin and flavor.
Certainly there are many who prefer pink tomatoes to red ones. I associate pink tomatoes with sweeter flavors and an absence of a certain umami "classic tomato" flavor, but whether that holds generally for genetic reasons or is just coincidental..? I did find some sweet reds that were in the class of pink associated taste too, so it's not 100%.
In the yellow vs white for clear skin, I've only grown and tasted one white tomato - that was Lotos - with a kind of unique flavor. Anyway no comparison possible. I have not grown enough GWR to give an opinion for them either. Yellow tomatoes have very divergent tastes. There are yellows that taste just like a red, you would not tell them apart in a blind test. Other yellows can be fruity, tangy, melony, or just mild. The fruity type of yellow is one of my favorites. Cherries like Galinas and Medovaya Kaplya.
There's a thread at OSSI talking about yellow, I've forgotten what the pigment is in "red" tomato skin but it is not the same as overall color chemistry, maybe it contributes to the taste. And there are different yellow alleles, so perhaps some are differently flavored? IDK.

[KaguyaCloud]

Perhaps not all flavor genes are linked with color, but maybe some flavor genes are linked with fruit color. For some reason, there seems to be quite a consistency in flavor types for cherry tomatoes that have red flesh, green flesh, and clear skin(Rosella, Reinhard Kraft's Purple Sugar, Black Cherry). This consistency is also to the point that crosses between Black Cherry and a clear skinned yellow in the dwarf tomato project lead to Dwarf Velvet Night(red flesh, green flesh, clear skin) and Dwarf Pink Opal(red flesh, clear skin) as the best products.

I'm now very interested in creating a purple micro with red flesh, green flesh, and clear skin if the flavors are consistent with certain fruiting traits.

[bower]

Black Cherry has some very dominant enjoyable taste traits. Lines from these crosses IME tend to be mostly great tasting.
Also some very dominant growth habit traits (huge plants, long internodes).

[Doffer]

For me F1 crosses with black opal where not special in taste. Indicating the taste genes of black opal are recessive.

[KaguyaCloud]

For example, it is very difficult to stabilize the superior taste of Sungold in a stable variety. Similarly, with Nebula, the F1 is incredibly flavorful, but in the F2, I only found about 10% super-sweet tomatoes with a Brix of 15.
This would suggest the presence of multiple QTLs that together are heterozygous for the good taste.

For me F1 crosses with black opal where not special in taste. Indicating the taste genes of black opal are recessive.

This is the reason that I am very weary of crossing with F1s. If the major flavors is coming from heterosis, then it might not be worth trying to create a homozygous open pollinated variety with it. Unless I have the space and time to get the best of the 4,096 homozygous combinations once it's stabilized, it's mostly up to chance.

Are there any known successful attempts in stabilizing an F1 variety? Recessive traits flavor traits from OP varieties might be beneficial since that could indicate certain flavors stabilizing within a line.

[bower]

A vast majority of F1's are bred for disease resistance and not especially for flavor. There may be a few cherries or saladettes F1s I've seen praised for flavor. So while there are doubtless F1s that have been grown out and stabilized for other useful traits, they are likely to be red tomatoes with classic flavors and not outstanding or unusual for taste.
When I'm growing out new crosses, I generally look for flavor in the F2 and ignore the F1 - mainly because final flavors are often masked by the F1 being red.

[Doffer]

A number of varieties with superior taste:
Sungold F1
Nebula F1
Picolo F1
Sungreen 4029 F1
Amelia F1

For most of them it seems very difficult to keep the taste when stabilization the variety.

[KaguyaCloud]

Update 12/5/24


Fruit set has already started. Overall density of the fruits is less than a typical micro dwarf, but more than the indeterminate parent. The high amounts of branching has lead to a very bushy growth habit. Both are around 6 inches(13cm) tall now.

The early stem termination gene definitely shortens overall height, but the nature of having 2 stems causes the axillary shoots to grow at 45 degree angles instead of horizontally. It definitely doesn't seem to have the smaller leaves of the micro parent, making it look more like the larger leaf micros that are around. I wonder if it's beneficial to prune to one stem, but I'm also curious to see how stable this strange genetic trait is.

[bower]

I can see benefits to the two stem trait. This allows the plant to balance its growth bilaterally. Also makes it easy for the tiny plant to ease its trusses toward the ground for support, if the set is heavy.

[KaguyaCloud]

Update 12/12/24


Very heavy fruit set from Joro F3-1(about 40) and rising amounts for Joro F3-2(around 30). The double stem and cherry tomato truss of 8-12 flowers really magnifies the potential number of fruits greatly. In addition, it seems that the flowers that bloom further down the truss shed significantly more pollen than the initial flowers did.

I also realized that it's likely the 2 locule trait will likely not pop up again, as the locule number(lc) is in the same chromosome as the dwarf trait(d).