[wrongwrong]

The problem as I see it, is that most of the traits in the canna world, aren't necessarily dominant or recessive, but rather co-dominant. And when working with only one side of the double helix, you are automatically excluding the possibility of those traits being present in the DiHaploid. This also ignores those traits that rely on more than one co-dominant gene, of which I believe there are many. Accepting that (to use your original example) the GS Cookies plant, is not a DiHaploid plant, but rather a diploid, then to try to replicate that plant using haploid cells will exclude half of the dna instruction sets necessary to create the plant.

I'm under the impression that it does not matter!, the plant in question can be heterozygous, therefore recombination and linkage is not an issue, so whilst "classical" breeding is all about homologous recombination this isn't. (but this is more to do with Reverse breeding, than what Spurr's initial course was/is.. where he planned to produce dihaploid progeny from DH parental lines).

Surely this sort of application is useful for putting elite's out in mass... and the artificial seed route would not be required.

[GMT]

Hmmmm, is classical breeding about that, or is it about getting a high proportion to exhibit desired traits? I don't feel they're the same thing at all.

[wrongwrong]

Originally Posted by GMT

Hmmmm, is classical breeding about that, or is it about getting a high proportion to exhibit desired traits? I don't feel they're the same thing at all.

Well if it was about getting a high proportion to show the desired traits you will not get to it faster than methods like this, selfing would be next... it would'nt be in a M/F method producing slow lines after lines taking years.. and even using M/F lines you would not be getting the homozygosity that you will here.... odds in your favour and time too... at the end of the day we want elite genotypes in our gardens showing us elite phenotypes.. (on a side note with homozygous lines/elite specimans etc you could find out what exact enviroment has an effect on any such plant, as such you'd know exactly how epigenetics effect the said plant).

Obviously for me it is impractical, I'm just a simple layman using my simple means which include selfing, I'm not getting a lab anyday soon... so I'll keep chasing my own tail... do you not think those that can incorp this type of tech should? I know you like all things "natural" but hey times they are a changing..

[GMT]

Well don't get me wrong, (sorry no pun intended), I'm not for one second suggesting that the Haploid plants wont show ANY desirable traits. I'm just saying that many of the traits being chased, wont be present in them if trying to replicate a diploid mother plant. You may well find many many new desirable traits, and the plants may well be very healthy, and yes, selfing a plant that started out as a haploid, will create a very stable offspring line, though femmed only, and so can never create a true breeding line without going through the supermale/male/male and female route, which would by definition ruin the goals. I know what the market likes at the moment, but what the market wants, and what's good for the future health and sustainability of the lines may not be the same thing. But that is a different concern. My original point raised, was that the process of meiosis throws the objectives into a little spin, and that by essentially throwing away one half of the pieces to the jigsaw, then completing the same jigsaw again becomes impossible. Though you may end up with a very nice little piece of the jigsaw that is much easier to complete time and time again.
Its way out of my reach too, and Mofeta clearly knows his stuff, and far more about this than I do, I just love these threads though, as they take the site beyond the typical "nice plant man" type of responses to read. Which I also post, and enjoy receiving but its nice to learn stuff too, and toss ideas around that makes everyone re-think things, and that's never a bad thing for anyone. As for should those who can, do, well, with a few provisions in place yes, of course they should. Namely that the work is seen in context, and that it's not the only work being done. But with so many pollen chuckers, breeders, and science bods out there all doing their own thing with our beloved plants, I'm sure we have the diversity of plants and methodologies to maintain a healthy population base.

[wrongwrong]

Its way out of my reach too, and Mofeta clearly knows his stuff, and far more about this than I do, I just love these threads though, as they take the site beyond the typical "nice plant man" type of responses to read. Which I also post, and enjoy receiving but its nice to learn stuff too, and toss ideas around that makes everyone re-think things, and that's never a bad thing for anyone.

I fully agree, I get bored with most things/threads on here, I need/want more interesting debate etc... though I need putting in place like we all do, I wish there was more top down feeding or such.. but then no one learns thru spoon feeding just more direction needed I guess.. spk soon.

[GMT]

man wrongwrong, that pm was an eyeopener, ok, I withdraw all objections/concerns I raised. With the right equipment, and knowledge, and chemicals, and protein solutions, yes, everything is perfectly possible.

[mofeta]

Hi everyone

Boy you guys are bringing this thread to life!

I happened to fall into a large job for someone unexpectedly yesterday, so I won't be able to post much for a while. I have some time to burn right now so I thought I'd jump in. I'll only be able to reply to about the first half of your posts, though.



Remember, I don't intend to do this technique, and I don't recommend it to anyone.

I just made the thread as an intellectual exercise because spurr was interested in discussing it. Also to illustrate to spurr that this was a LOT of work, and had limited context in which it was useful. Especially the centromere mediated technique.

Gamete culture could be done by a serious ameteur at home (with a lot of work), but the centromere stuff is professional lab/expert technician stuff only. This will be (maybe already is) done by big professional outfits like GW. Not by me though. The standard breeding techniques produce results that satisfy me.

That being said, on to discussion.

Originally Posted by GMT

I feel it would become impossible to replicate the traits expressed within the original GSC given as I said earlier, that many traits will not only be co-dominant, but also rely upon multiple genes that are all co-dominant and essential to all be present for the particular trait to manifest. In this sense, no haploid can actually be a full instruction set of the parent, even when doubled

I totally understand your point, and it is indeed a good question, sorry I didn't make that clear. (As a note, the term polygenic is a useful term to describe traits controlled by multiple genes. Also, instead of "co-dominant" you should say "dependent on" or "influenced by". The term co-dominant has a very specific, precise meaning. It is indeed one of the ways that genes will effect each other, but there are a lot of ways that genes influence each other.)

Here is why I think it doesn't matter. It is the process you brought up a few posts back- meiosis. That is why I wanted to explore DNA and chromosomes, because to understand meiosis properly, you have to have a good picture in your mind of these things.

The process of meiosis produces monoploid haploid gametes, the one set of chromosomes each of the four gametes gets is a mixture of the two sets that the plant producing the gametes got from it's parents. Some of them (maybe very, very few) would happen to have the right combination of alleles to show an excellent Cookie phenotype. The rest are culled. There is more to it, and nunsacred's comment is perfect lead in:

Originally Posted by Nunsacred

Someone said earlier that a trait might be only available if the two parental contributions are both present, which is true, this might happen quite a lot.
So DH populations might entirely lack the desired trait.

^^ This is true.
Think about it, it's more likely to be just 'general healthiness' that gets 'broken' by losing a haplotype, more likely to be a struggle to find a healthy enough phenotypes there.

That's a feature, not a bug. The perfect homozygousity of the DH plants exposes deleterious recessive alleles, allowing for culling.

Remember, this technique requires good selection from large numbers.

I cannot think of a reason why a diploid obtained from doubling a haploid would be different than a regular plant, except for the perfect homozygousity.

Think about incrossing plants, like tomatoes. Open pollinated varieties are easy to maintain over many many generations because the are so homozygous. They got that way because the seeds in a tomato fruit are all from pollen from the same plant- selfed. They exclude pollen from other individuals by flower morphology- the female parts never grow far enough out to get past and out of the male parts of the same flower (with most cultivated varieties anyway). To get hybrids you have to remove the seed parent's male parts early, and hand pollinate with pollen from another plant.

How can a plant that only selfs have vigor (like all the open pollinated tomatoes I've ever grown, which is a LOT) you ask? It is because the deleterious allele combos have been weeded out, by virtue of being exposed by selfing.

Tom Hill has made posts here somewhere (and I paraphrase loosely) that if he were to undertake making seed lines from elite cuts, he would treat the cut like and incrosser, selfing and such until he had very homozygous lines, culled of deleterious alleles.

Sure, real high numbers of the first few generations would be pretty weak, but some (again, maybe very very few) would be what you want and you use those. This technique is not preservation, it is cut throat inbreeding- purposeful bottlenecking.

Say you used standard breeding techniques on your favorite line of weed. After years of work you have a F20 line that is killer smoke, vigorous and real homozygous- a pack of seeds would come out very similar ("like clones" is a description I have seen in seed catalogs for some strains).

How is this different from a doubled haploid (excluding the perfection of the homozygousity of the DH)?

Boy, I wish I had more time, because the posts you guys made after the ones I just replied to are quite thoughtful, and really do get into the meat of the matter, you all make good points.

Due to the work that fell into my lap, by the next time I can check in you guys will probably be posting pics of you DH lines!

[GMT]

Hey mofeta, well because while many of the genes that we want will be at diff loci, some wont be, and then these become an either or situation unless we can move their loci in some controllable way. In the example you give, in each generation is the possibility to have both ( and in many cases as we have agreed, many) genes present without that either or situation arising. Therefore after many generations, there will be many fine examples of the original within numbers. For the DH tech to work, it would be necessary to marry one set, with the opposing set, which with numbers, genetic testing equip, and the ability to prevent recombination at meiosis, can, it now appears, be done. The resulting offspring being a perfect copy every time. Therefore the DH tech, with the right knowledge and access, is the far superior one. Though I don't know how much the beans would cost after all that.

[GMT]

Hi Tom, Is my maths out or have I got this right:
each DH has 46 combos (assuming recombination is stopped)
But each one (without dna testing) may need 95 crosses to find the perfect match assuming no duplications and 96 seeds to work with.
Making 1% ish the perfect result.
Of course after that its plain sailing to perfect replicas every time (again providing recombination is always prevented.

[wrongwrong]

Originally Posted by TomHill

Hi GMT yeah 1% may be a more accurate figure. I have not crunched the probability maths so much other than to be excited about the relatively low numbers required. I believe 98 brings P=99%. Here is a handy chart https://www.ncbi.nlm.nih.gov/pmc/arti...05/table/tbl1/ from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2784905/

is that R dirks/wijnklers paper? have to run now... I'm guessing it is haha