Breeding Sweet Potatoes (turning them into a seed grown annual)

[reed]

Really interesting discussion. I really like the main goal of breeding for fast tuber growth and seeds in one season as opposed to for cloning like was mentioned. I wonder if the same could be bred for in regular potatoes with TPS.

I sure wish someone would do it with TPS and maybe throw in a little more heat and drought tolerance to go along with it. I would definitely trade some TSPS for them.

It was actually pretty easy to do it with sweet potatoes. It was just that at the start, not all of them did it. Back then only maybe 10% - 15% did all four of the necessary things, 1-sprout easily from seed, 2-make big storage roots, 3-bloom a lot and 4-make seeds. That number has improved a lot over time. I think it's well over 50% now, but I would like to see it confidently measured at 90%.

[reed]

I have transplanted 150 seedlings into medium sized pots where I hope to be able to determine things like whether or not they have the clump root trait and what type of vine growth they have. I may start culling the very large vines fairly early or at least keep them heavily trimmed to keep them form overrunning the place. Also put five seedlings in each of ten large tubs so 200 planted so far. That accounts for about half of the total and I'm running out of pots.

I tagged about 15 to leave where they sprouted, the only ones I'll be able to say were truly direct planted and the seed beds were pretty small so they might be crowded later on, especially if they turn out to be large vine types. 

Already with this amount sprouted I can see that the ones with dark purple cotyledons are about 15% of the total. Lots of other variations though, different shades of green, green with purple or red stems and so on. Can't tell yet except of a very few what the other leaves will look like but the few so far are all the plain non-lobed kind and light to medium green. Maybe 10 out of the bunch had the white, pink or yellow cotyledons. When the cool cloudy days switched to the hot sunny days, they all died. 

Two seedlings have four rather than two cotyledons, three seedlings have three cotyledons. They are all very dark purple and very small. They don't look unhealthy. The stems are just as thick as the others or even thicker, but the leaves and the height are only about a third the size of the rest.  I'm very intrigued by these dark purple "miniatures" and intend to keep a close eye on them.

[AndrewB]

Your potted TSPS sounds similar to what I'm doing with TPS this year. I've got a pot of many TPS sprouts all growing. I've noticed several are growing much faster and bigger than the rest. I have been trimming them to try and let some of the others catch up. In my case i planted tetraploids and diploids together, so it is possible that the fast large ones are tetraploidy. In any case I will evaluate them by biggest tubers in the first season and see if any are worth attempting to save for next year. If any make TPS in this first year I will also try to save those (assuming they also have tubers). I suppose one should be careful saving TPS from plants that might produce TPS in the first year, but little to no tubers.

[reed]

Your potted TSPS sounds similar to what I'm doing with TPS this year. I've got a pot of many TPS sprouts all growing. I've noticed several are growing much faster and bigger than the rest. I have been trimming them to try and let some of the others catch up

Sweet potatoes come in so many different phenotypes as far as how large the vines get, long, bushy or semi-bushy. I hope with such a large number of them this year to see if I can learn anything about predicting eventual growth at an early stage so that in later seasons, I'll know which ones to cull early on.  Already, I'm starting to suspect those getting a bit taller and with longer internodes might be the large vine types. I can't cull them though because I won't have any idea of how they are at blooming for a good while longer. I do anticipate keeping them clipped back if they get too out of hand.

In my case i planted tetraploids and diploids together, so it is possible that the fast large ones are tetraploidy.

I don't completely understand ploidy levels, although I did come across it in researching about sweet potatoes. But there are no references or information that I found that would indicate which ones might be which so even if I researched it better, I'd still be mostly in the dark as to how it might be at play in my project. If I remember right, they do vary for it but that it isn't the culprit in their weird compatibility issue. It sounded like that was more related to the physical size and structure of the pollen and the pollen tubes, but I don't remember those issues and ploidy levels being related. It's one of those questions I have little chance of finding an answer to. Fortunately, I don't have to. They grow and bloom and make seeds and roots or they don't, even if I don't understand exactly why.  I do wonder if the ploidy could levels have something to do with the ones with anomalous numbers of cotyledons. 

In any case I will evaluate them by biggest tubers in the first season and see if any are worth attempting to save for next year. If any make TPS in this first year I will also try to save those (assuming they also have tubers). I suppose one should be careful saving TPS from plants that might produce TPS in the first year, but little to no tubers.

In the first years with sweet potatoes I was just happy to get seeds. Less than half back then made good sized storage roots and still not all do. Potatoes and sweet potatoes are such different critters, about the only thing in common is they both grow in the ground. Sweet potatoes are roots rather than tubers although I don't completely understand the difference. Except that if you plant a whole sweet potato or even half of a sweet potato (rather than a slip) it just starts growing again rather than rotting away like a seed potato does. 

[nathanp]

Quote from: Andrew Barney on 2022-05-31, 02:51:41 PM

    In my case i planted tetraploids and diploids together, so it is possible that the fast large ones are tetraploidy.


I don't completely understand ploidy levels, although I did come across it in researching about sweet potatoes. But there are no references or information that I found that would indicate which ones might be which so even if I researched it better, I'd still be mostly in the dark as to how it might be at play in my project. If I remember right, they do vary for it but that it isn't the culprit in their weird compatibility issue. It sounded like that was more related to the physical size and structure of the pollen and the pollen tubes, but I don't remember those issues and ploidy levels being related. It's one of those questions I have little chance of finding an answer to. Fortunately, I don't have to. They grow and bloom and make seeds and roots or they don't, even if I don't understand exactly why.  I do wonder if the ploidy could levels have something to do with the ones with anomalous numbers of cotyledons.

Sweet potatoes are hexaploid and appear to be much more complex because of that, making them very different from other crops, like potatoes, which are largely tetraploid. 

Technically, I believe they would be considered allopolyploids (is that correct)?
I believe some of the problem with the compatibility issues comes from this, meaning parts of the polyploid genome are compatible with some other sweet potatoes, while parts are not compatible.  There would be tetraploid part of the hexaploid genome, and a diploid part, coming most likely from two different species ancestors.  Or something like that.  The genetics got mashed together into the hexaploid genome.  And the tetraploid part is only compatible with some other sweet potatoes' tetraploid genes, while not compatible with others.  Same with the diploid part.  And the diploid and tetraploid parts would not be compatible with each other.  Or something like that.  I'm no geneticist, so I'm sure I'm not 100% accurate with this.

[AndrewB]

My understanding is that tetraploids and higher ploidy numbers are generally bigger and perhaps faster growing because the growth genes are more or less doubled or more as well. Thus they basically get double the hormones.

This is why tetraploid potatoes are bigger and why modern strawberries are much larger than wild type mountain strawberries.

[reed]

Sweet potatoes are hexaploid and appear to be much more complex because of that, making them very different from other crops, like potatoes, which are largely tetraploid. 

Technically, I believe they would be considered allopolyploids (is that correct)?
I believe some of the problem with the compatibility issues comes from this, meaning parts of the polyploid genome are compatible with some other sweet potatoes, while parts are not compatible.  There would be tetraploid part of the hexaploid genome, and a diploid part, coming most likely from two different species ancestors.  Or something like that.  The genetics got mashed together into the hexaploid genome.  And the tetraploid part is only compatible with some other sweet potatoes' tetraploid genes, while not compatible with others.  Same with the diploid part.  And the diploid and tetraploid parts would not be compatible with each other.  Or something like that.  I'm no geneticist, so I'm sure I'm not 100% accurate with this.

I had to look up allopolyploids and I think your guess is probably better than mine as to how it affects compatibility.  I also do remember in my research, coming across them having a two species origin. As I recall they still don't really know what two species were involved. They also don't know when, but believe that it happened pretty recently, maybe just several thousand years ago, probably in south or central America.

I also remember something about them being naturally transgenic, having picked up a bit of genetic material from an unrelated species and that was even more recent, just a few thousand years ago. I don't remember that being implicated in the compatibility issues nor their habit of spontaneous mutation, but it might have something to do with the sugars they make and store in the roots.

Mutation in clones is most common in the first few years after a new one germinates.  They can revert back to the phenotype of one of the parents, usually the mother or they can mutate to something completely different. I didn't really believe that until I saw slips from the same root develop two different colors in their roots. I don't know if I've had any revert to the mother type, because I don't keep good enough track but the two from one stood out. And I guess if the root color can mutate about anything else can too, including compatibility. 

[nathanp]

The unbalanced hexaploid genome separates how they behave in breeding, differently than potatoes, which are mostly tetraploid.  Tetraploid potatoes will cross and neatly result in crosses which have two alleles from each parent, in nearly all cases (there are exceptions). With the sweet potatoes hexaploid genome, there is an uneven split.  One parent needs to contribute a full tetraploid set of alleles, and the other a diploid set.  But which one does which, and in what possible combinations?  The incompatibilities would have to do with which combinations will be compatible or not compatible with other donors.  And if there are at least two species involved in their origin, the compatibility of each may still be dictating what is happening at some level.

[reed]

In other words, they are pretty screwy. It goes along with what you said earlier about the compatibility groups. I potted up a bunch more yesterday and was sitting there thinking about this. I suppose I could take a single plant, assuming it turns out to be one that blooms at all. Then, carefully pollinate individual flowers one at a time, with pollen from all the other individual plants, taking all the necessary steps to make sure that each flower wasn't accidentally otherwise pollinated. And keeping track of which other plant provided the pollen for each individual flower.

I don't remember statistics well enough but even if I did that with just one plant, and with just 99 other plants providing pollen, the task would be enormous. To do it with each of the 100 plants, matched to each of the other 99 plants would be absolutely ridiculous. Trying to do it with all of the approximately 400 plants would just be crazy.

Do you think it's possible a universal compatibility might exist? That is, a plant that was compatible with any other plant.

[add]
What about the ones with three, four or five sets cotyledons? Or the ones that grow their first true leaves from a branch below the cotyledons, or the ones that apparently can't make chlorophyl? Are those things related to the unbalanced hexaploidy? 

If these compatibility groups exist as it sounds like they do, I wonder which or how many of them I have. I have seen examples of plants cloned from the same root make lots of seed in one part of the patch and few seeds in another part. I guess the seedy ones had immediate neighbors that were compatible, and the non-seed ones didn't.

Last year or the year before I had two volunteers that came up about six inches apart and formed a comingled mass about four feet around. Nearly all the flowers in the very comingled part made seeds but the edges of both plants made noticeably less seeds. I think that's because since the bees are very methodical, moving from one flower to the next closest mixed them up more where the flowers were close together, whereas the flowers on the edges were more self-pollinated and the individual plants were apparently not self-compatible.

I have gotten lots of seeds from as few as twenty plants, but I still have a feeling that there may be a minimum number needed to ensure that some will end up being compatible. Especially if you don't have any clones of know compatible ones from the previous year.

Not accounting for those that in fact are self-compatible. Those at least are fairly easy to test for. Just wait till they start blooming and root a cutting, isolate it and hand pollinate. I've done that a little bit already, maybe I'll do a lot more of it this year. Even though I'm not really100% sure that self-compatibility is necessarily inherited by the next generation.

[Tim DH]

Hi Nathan,
   I'm curious about your suggestion that the hexaploid genome of Sweet Potatoes is split unevenly at meiosis 4/2. (Its not a plant I have much experience of.) I have an orchid collection which includes diploids, triploids, tetraploids, pentaploids and hexaploids. My assumption with the hexaploids was that they would always split 3/3.

   Wouldn't the existence of tetraploid gametes result in the possibility of Octoploid plants?

Tim DH

[nathanp]

Hi Nathan,
   I'm curious about your suggestion that the hexaploid genome of Sweet Potatoes is split unevenly at meiosis 4/2. (Its not a plant I have much experience of.) I have an orchid collection which includes diploids, triploids, tetraploids, pentaploids and hexaploids. My assumption with the hexaploids was that they would always split 3/3.

   Wouldn't the existence of tetraploid gametes result in the possibility of Octoploid plants?

Tim DH

Good question, and I don't really know.  But this paper is well worth reading if you want to sort through this.  I believe this implies that the B1B1 diploid chromosome pair crosses with the B1B1 of another sweet potato, while the B2B2B2B2 tetraploid chromosome set crosses with the tetraploid B2B2B2B2 of the the other.  Or I really don't understand this at all.

In others words, I. trifida would be a direct progenitor to only one (Bs1'Bs1') of the constituent diploid subgenomes in sweetpotato. The other two subgenomes in sweetpotato would have been derived from hybrids of two pairs of duplicated diploid genomes (in Ip-Tn and Ip-Lt) in a respective niece and aunt kinship to that of I. trifida. They would thus be extremely closely related to, but not the same origin as the first sweetpotato subgenome from I. trifida. Additionally, I. tenuissima and I. trifida would be in a niece and aunt kinship, derived from the predicted ancestor Ip-N2 hosting the network nodes N2 and M, and in a third and second cousin kinship to I. littoralis, respectively. By the parsimony principle, I. littoralis was modeled as a paleo-tetraploid derived from the predicted Ip-Lt tetraploid. Since the three diploid subgenomes in sweetpotato were only partly differentiated as represented by the three inter-subgenomic pairs of differentiable lineages of some COSII genes, they were thus noted using the three color-gradients blocks under the subgenome designations.

https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0229624

[reed]

Does multiple subgenomes mean that there are different sweet potatoes or that there are three different genomes within a single sweet potato?

[nathanp]

Does multiple subgenomes mean that there are different sweet potatoes or that there are three different genomes within a single sweet potato?

Yes, there are three different genomes.  Two from one species that represent a tetraploid component, and one from another species that is a diploid component.  That paper in the link above goes into the details. 

[reed]

I've skimmed over the linked paper but have to admit my education level related to genetics is a bit lacking. Makes it rather tough to chew on although fascinating at the same time.

This sentence in the introduction amuses me a little bit, "A more effective genetic improvement of sweetpotato depends on advances in our understanding of its genome and genetic inheritance." I guess that's so but from what I gather from the North Carolina and Louisiana universities over the last ten years they have started around 800,000 seedlings.

In that same time, I have started about .0025% of that and I've seen all kinds of interesting things show up. They have lots of staff and other resources, why don't they just observe and select like I do? 

[reed]

Transplanting more seedlings last evening, some good indications as far as determining some traits very early on. Those not yet transplanted, while still pretty small have grown enough that I can tell what color the roots will be, at least on the outside. Also seems fairly apparent which will be bushy with some having shorter internodes and already developing a branching habit. Have half a dozen or so are already developing bloom buds even though they only have a few leaves, going to keep a close eye on them over the season.

Probably 2/3 of the bushy ones have the deeply lobed leaves and all the ones with bloom buds have the heart shaped leaves but I don't know if that really means much of anything. I've lost some very dark purple ones to some critter ; I suspect it is birds because some of them were well above the ground in the big tubs.

I worry about how much it will set them back as far as root development but in another couple of weeks I think I'll apply the LSU method of dumping them out of their pots or digging them up to see what I've got. I hope this year to find a good number of seedy, bushy ones with nice, tasty roots that I can clone next year.