Doubled haploids

[mofeta]

Hi guys

Thanks for exploring this with me! Your posts are very thoughtful and stimulating. I have been real busy lately, sorry I haven't contributed much. I think about it a lot though. If I was not constrained by my "real world" obligations I would obsessively study this stuff.

These are deep waters, indeed, and my understanding of them is woefully incomplete. I need to read a LOT of stuff to really fully grasp these exciting new developments in plant breeding.


The paper posted by Tom (Dirks 2009), and the corresponding patent quoted by wrongwrong are quite interesting. I suggest careful study of them. They are a great springboard from the discussion of DH to the broader concepts of reverse breeding.

GMT- now I understand what you were trying to get at with your comments about meiosis. You were talking about reverse breeding as opposed to DH. I was talking about using DH as a shortcut in a breeding scheme, you were talking about making EXACT REPRODUCTIONS of a heterozygous individual (reverse breeding). When I talked about avoiding meiotic shuffling, I meant in a DH context (crossover of perfectly homozygous homologs is basically the same as no crossover). If it is not clear to anyone that DH and RB are different things, we should discuss that.

As the Dirks 2009 paper, and the patent quoted by wrongwrong, lays out, for RB, you must suppress crossover by inducing achiasmatic meiosis. The Table 1 that you guys are referencing from Dirks 2009 is for non-recombinant DHs.

In Dirk's RB tech, you would use the scheme laid out in Figure 1 of the 2009 paper for GSC, as it is an unknown heterozygote. I think it would probably be easier (HA!) to use the apomixis tech from the second paper in my cenh3null thread: Synthetic Clonal Reproduction Through Seeds


If anyone has access to a university account, it would be interesting to see the full text of this paper:

Reverse plant breeding success


I should mention that where DH is non-trivial for an amateur, RB is an order of magnitude or two more difficult. Pro lab stuff exclusively.

Man, I wish I had more time. I could write a couple thousand words replying to what you have already posted. Maybe tomorrow I can delve further into this stuff with you guys (or, if you post some irresistible stuff now, I will unwisely refrain from sleep tonight!)

Thanks again, everyone. I really appreciate the depth you have brought to this thread.




PS: Hey wrongwrong, is your handle an allusion to the brilliant Eno/Quiet Sun/801 Live song? The version on 801 Live (one of my all-time favorite recordings) is killer.

I'm looking in my little black book
To see if I was right or rongwrong
Between the lines on the tattered pages
My spiderly writing inclines
I'm old before my time
I feel that I'm growing out of this world
But with the world at my ears
I guess it's true there's no tears, no tears
When things get bad I can always turn into a cloud then
I'll drift back home if the wind will blow me there

 

[GMT]

accepted, I do see this as being more important in the preservation, or rather replication of stunning individuals rather than as a tool to produce identical seeds to a plant as yet unknown. I don't agree though that meiosis in a dh is the same as preventing the re-writes that take place. though I don't know if I'm over valuing the process or if you're undervaluing it.

 

[wrongwrong]

@ Mofeta, on synthetic clonal reproduction, heres the patent for it

http://patentscope.wipo.int/search/...=&sortOption=&queryString=&tab=PCTDescription

can't seem to get a hold of the paper itself!..

if it was the apomixis route, you'd need irradiated pollen too ( I think)

http://www.niser.ac.in/~cellbio2011/abstracts/Abstract-Imran Siddiqi.pdf

 

[wrongwrong]

Kopite, closer to strawberries than corn, ie we'll never get there (not where we are talking about) with traditional conventional breeding practices imo. I also believe I read the material correctly and that 98 outputs are what we are talking about, not so deep a search, as marveled by others (in the link I posted) as well as me.

Strawberries, are bred to find stand out individuals are they not? so they use a large number of plants? BUT they usually use a good stock to start with? so without the good stock at the beginning you have more work!.. but then as I understand it you get at it fast, select the best 5% say and ONLY select the best of the best from them... gets us to elite x elite thinking... now that was a good discussion on CW with hyb and Vic more or less on these lines of thinking.. to me you need both approaches.. or are we more or less fully in balls/elbows deep with the DC pool and should just be cutting to the chase and forgetting all the fore play? After all how many of the stand outs in the community are from random crosses?

 

[wrongwrong]

PS: Hey wrongwrong, is your handle an allusion to the brilliant Eno/Quiet Sun/801 Live song? The version on 801 Live (one of my all-time favorite recordings) is killer.

my dad is a big roxy music and eno fan, we used to have oldfields tubular bells on all the time... so yes though I got the name wrong..

Oh, oh, oh, oh, oh, oh, oh, oh, here we go!
We are the table the captain's table let's get it understood
Let's get it understood
We are the losers we are the cruisers let's get it understood
Let's get it understood
We are the diners the final diners let's get it understood
Let's get it understood

Most of us are tinkers, some of us tailors
And we've got candlesticks and lots of cocktail sticks
We saw the lovers the modern lovers and they looked very good
They looked as if they could
We are the neighbors the nosy neighbors we think just like you would
We think just like you should.

sorry for the side track.

 

[wrongwrong]

hmmm, before we get to counting stuff, does anyone know if what is required to prevent recombination in canna, has been identified and is available? I know for a bunch of stuff it has, but if its not done for canna, then the numbers are going to get much bigger.
Also where is the 98 coming from? I'm still only counting 96, seem to be missing 2 somewhere.

Well I only got to the 1st part of it, 2x/(2x)2 = (½)x and 1−(½)x = (2x−1)/2x

1024/1048576 = 0.0009765625

so then

1-0.0009765625 = 0.9990234375

so figuring the probability is 0.99 I then just looked at chart..

after the chart, got caught on this bit!

Reconstruction of heterozygous germplasm

For crops where an extensive collection of breeding lines is still lacking, RB can accelerate the development of varieties. In these crops, superior heterozygous plants can be propagated without prior knowledge of their genetic constitution (also see Figure 1). Table 1 shows the number of doubled haploid plants that are necessary to reconstruct the starting plant at different levels of probability. The number of DHs that is required is surprisingly low. For instance in maize (x = 10) just 98 DHs are expected to contain a set of two reciprocal genotypes (P = 99%).

Breeding on the single chromosome level

Many interesting characteristics in crops are based on polygenic gene interactions, very often located on different chromosomes. These quantitative traits are therefore not easy to breed on. Figure 2 explains how chromosome substitution lines can be obtained when RB is applied to an F1 hybrid of known parents. These homozygous chromosome substitution lines provide novel tools for the study of gene interactions. When crossed with one of the original parents, hybrids can be formed in which one of the chromosomes is homozygous (Figure 2, lower right), whereas it is also possible to produce hybrids in which just one chromosome is heterozygous (Figure 2, lower left). The former allows the study of epistatic interactions between the background and genes contributed by the substitution chromosome. Offspring of plants in which just one chromosome is heterozygous, will segregate for traits present on that chromosome only. Selfing plants that carry a substituted chromosome (or using recurrent backcrosses) will allow breeders to fine-tune interesting characteristics on a single chromosome scale. This could bring forth improved breeding lines carrying introgressed traits.

 

[mofeta]

Hi wrongwrong

Don't miss the fact that the DHs they are talking about are non-recombinant, produced from gametes in which crossover was prevented, ie achiasmatic meiosis.

EDIT: I went back and reread your and the others' posts. You guys obviously have a good handle on this, I apologize for not reading your stuff with the care it deserved. I thought you were missing the non-recombinant part on that Table 1.

Sorry to all, I will be more careful from now on. When I get a chance I will try to post something useful for a change.

 

[GMT]

Well I only got to the 1st part of it, 2x/(2x)2 = (½)x

yes

and 1−(½)x = (2x−1)/2x

er..... well if X equals 1 then yes, but then why use algebra at all?

1024/1048576 = 0.0009765625

so then

1-0.0009765625 = 0.9990234375

so figuring the probability is 0.99 I then just looked at chart..

after the chart, got caught on this bit!

ok, where did they come from? lol I'm off to do a diagram, we need some pics for this, I may be gone some time

 

[GMT]

lol, right ok got your numbers, =1/(0.5*0.5*0.5*0.5*0.5*5*0.5*0.5*5*0.5)*100 =1024
so there are 1024 possible haploid combos from one diploid donor plant. But that's mathematically, 512 perfect pairs capable of reproducing the donor. So 1 is reversed etc, and 1023 are pollinated, hoping that your 1024 are all different (which of course they wont be, and hoping that the perfect pair to the one you picked is in the rest.)

 

[mofeta]

More stuff to check out:

Plant Breeding Revised and Reversed
This a PDF of the visual aids from a presentation at a conference in China a couple of years ago. It recapitulates the paper being discussed, has some cool photos, and has a lot of other interesting stuff.



NEAR REVERSE BREEDING

I'm not going to let myself read this one until I have nailed down the reverse breeding calculations.

 

[Tom Hill]

this is fun guys, thank you. I can hardly keep up.

Kopite, what I am saying and have been for some time (re strawberries vs cannabis) is that the octoploid (in most indiviaduals) strawberry, will come to be known as no more/less complicated than drug cannabis (recombinations of terpenes for starters), and that in the continued absence of asexual reproduction, it (cannabis) should be treated (while we review, and reconstruct it's breeding protocol and this is what I/we are contemplating) differently, much more as strawberry is than maize. We have had this conversation before. Yes, to reach valuable combinations amongst strawberries we combine favored candivars, but it's from that point onward where we really make progress is it not (?) with identical reproduction and thus, this thread is interesting regardless of where we are with the tech.

 

[Tom Hill]

re reverse breeding " The number of DHs that is required is surprisingly low. " is the consensus.

 

[Tonygreen]

I just want to say I admire you guys for your tenacity and drive, I'll be off smoking in a corner, let me know when you get it figured out.

 

[wrongwrong]

this is fun guys, thank you. I can hardly keep up.

Kopite, what I am saying and have been for some time (re strawberries vs cannabis) is that the octoploid (in most indiviaduals) strawberry, will come to be known as no more/less complicated than drug cannabis (recombinations of terpenes for starters), and that in the continued absence of asexual reproduction, it (cannabis) should be treated (while we review, and reconstruct it's breeding protocol and this is what I/we are contemplating) differently, much more as strawberry is than maize. We have had this conversation before. Yes, to reach valuable combinations amongst strawberries we combine favored candivars, but it's from that point onward where we really make progress is it not (?) with identical reproduction and thus, this thread is interesting regardless of where we are with the tech.

I think I agree, I guess it's like mofeta put somewhere - it's not how you find the gold or even where, just as long as you find some etc etc

(ps - either you're up late, or I'm early!)

 

[Tom Hill]

have you guys not seen each other before? I pulled all of your covers in those who I enjoy reading in the my favorite breeders thread, ahahaha. Soon we'll yank Hyb's covers too, and get Charle's X into the mix, all crazy fucks of various knowledge who have one thing in common. They are not afraid to dive in.

 

[GMT]

For creation the numbers may be lowish, but for replication they are staggeringly high. The 98 number, from what I can tell is the number for creation, the 1024 is for replication. But is the 98, 2 of them will be suitable companions, rather than no matter which you pick, one of the other 97 will be good for it? Because if its the former, thats a lot of crosses needed without lab equip.

 

[mofeta]

Hi Tom

I didn't know you were thinking of actually doing RB. Very ambitious. The resources needed really are staggering to contemplate (for me anyway). I'll have to shift my focus to thinking about practical application along with the theory. I'm all for it!

Hi GMT

Yeah, you are correct. Although you only need to successfully create 98 non-recombinant DHs to be fairly certain that 2 of them are complimentary parents that will result in recreating the desired elite heterozygote, figuring out which two would be some work. I guess that on the bright side, you would only have to grow one seed from each test mating.

That would be relatively trivial though, compared to the high tech needed to produce the non-recombinant gametes in the first place, as you have pointed out a number of times.

 

[GMT]

Yeah that's how I read it too mofeta, but I still don't think that finding the right 2 out of 98, ie 1x97, 98 times, plus one seed from each mating, can be done by a private breeder. Ignoring the tech, simply the space, organisation and time make the price of any seeds produced, outside of the standard commercial seed price range that customers will pay for. Remember, to most of us, making a copy of a plant, just takes a pair of scissors and some spare compost . And each seed planted is a lotto ticket, I think this is purely for the pharma market place, where replication of a standardised product is essential. I got excited at first, thinking this may be a short cut to me stabalising my tri line, but alas, the work continues.

 

[mofeta]

....finding the right 2 out of 98, ie 1x97, 98 times....

It's not that bad. Since reciprocal pairs would be equivalent, there would be 4753 combinations.

To find the number of pairs from n parents, where order doesn't matter:

n(n-1)/2

 

[mofeta]

Really, the nut of the issue if you want to realize the RB scheme of Dirks is the production of the non-recombinant gametes by inhibiting crossover.

From the paper, of the routes proposed for this, the only one of any utility for anyone other than cutting edge professional labs is the exogenous chemical one. The paper mentions this compound as a possibility:

mirin

It seems that mirin inhibits the MRN complex (Mre11, Rad50 and Nbs1). Apparently, the MRN complex is essential for the initiation of the DSBs (double strand breaks) necessary for crossover in meiosis:

Meiotic DNA Double-Strand Break Repair Requires Two Nucleases, MRN and Ctp1, To Produce a Single Size Class of Rec12 (Spo11)-Oligonucleotide Complexes

Say you bought some mirin from the Cayman Chemicals link I gave above. You then treat your elite heterozygote with it. How would you know it worked?

You would either have to sequence a number of individual gametes produced by the plant, or by actually using the gametes in a RB scheme and seeing if it worked.

The former would require a high tech lab and people that knew how to use it.

The latter would require prohibitive amounts of work and time: for each experiment to see what route of administration, concentration etc etc, actually worked (if at all) you would have to carry out a full RB growout and see if you got what you wanted.

I think this is one of those cases where we have to wait for the pros to work it out.