Author Topic: Marker Assisted Selection, Promiscuous Tomatoes  (Read 170 times)

Joseph Lofthouse

  • Administrator
  • Full Member
  • *****
  • Posts: 342
  • Karma: 44
  • Great Basin desert, Rocky Mountains
    • Open Source Plant Breeding Forum, founder. World Tomato Society, ambassador. Plant Breeder. Yogi. Shaman.
    • View Profile
    • Garden.Lofthouse.com
    • Email
  • Koppen zone: Dsa
  • Hardiness Zone: USDA Zone 5
Marker Assisted Selection, Promiscuous Tomatoes
« on: 2021-04-05, 01:35:13 PM »
Someone offered to do marker-assisted selection on the promiscuous tomatoes. The DNA-sequences are known. It's the HT allele that's broken in domestic tomatoes. Some of the base pairs are out of order.

I'd sure like to accept the offer, but the growing season is upon me, and I'm still spending tons of time on the book, so it's hard to do the research.
« Last Edit: 2021-04-06, 11:33:44 AM by Joseph Lofthouse »

William S.

  • Full Member
  • *****
  • Posts: 820
  • Karma: 47
    • Botanist, gardener, and preservice science teacher.
    • View Profile
  • Koppen zone: Dfb Googled
  • Hardiness Zone: USDA zone 6A
Re: Marker Assisted Selection
« Reply #1 on: 2021-04-05, 02:58:48 PM »
Someone offered to do marker-assisted selection on the promiscuous tomatoes. I believe that the allele-sequences are known. I think it's the HT gene that's broken in domestic tomatoes.

I'd sure like to accept the offer, but the growing season is upon me, and I'm still spending tons of time on the book, so it's hard to do the research.

That sounds like a lovely offer. What's involved? Could we accept it as a breeding team?

I have a couple trays of promiscuous project seedlings (85 plants in three accessions: bicolor 2020=C31 XA1/A2 3 species, R18, & S35-7). I have plenty of the bicolors I selected from C31 XA1/A2 and could plant however many are needed. My plan was to direct seed some of them. I could in theory do something like number each seedling, clip a leaf off each of them, and send the corresponding numbered envelopes with leaf samples in to said researcher. The researcher could ID the seedlings lacking the broken HT allele. Then I could plant the marker assisted selected seedlings in an isolation field 150' from other tomatoes.
« Last Edit: 2021-04-05, 04:24:33 PM by William S. »
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

Joseph Lofthouse

  • Administrator
  • Full Member
  • *****
  • Posts: 342
  • Karma: 44
  • Great Basin desert, Rocky Mountains
    • Open Source Plant Breeding Forum, founder. World Tomato Society, ambassador. Plant Breeder. Yogi. Shaman.
    • View Profile
    • Garden.Lofthouse.com
    • Email
  • Koppen zone: Dsa
  • Hardiness Zone: USDA Zone 5
Re: Marker Assisted Selection
« Reply #2 on: 2021-04-06, 12:16:14 AM »
The physical work is easy. Collect 128 leaf samples, 6 mm in diameter,  put them into micro-centrifuge tubes, already in their holder. Add a few controls, and duplicates. And record everything!!!! ha!

The hard part is knowing what to ask for, and understanding the genetics.

I think that these articles might be helpful.

https://onlinelibrary.wiley.com/doi/10.1111/tpj.12424
https://nph.onlinelibrary.wiley.com/doi/full/10.1111/nph.14130
https://www.academia.edu/14032701/Marker_Assisted_Selection_in_Tomato_Breeding

I think that this is the broken gene:
https://www.ncbi.nlm.nih.gov/nuccore/AB072478

I should get hold of the people doing the actual work, and have a conversation.
« Last Edit: 2021-04-06, 12:21:17 AM by Joseph Lofthouse »

William S.

  • Full Member
  • *****
  • Posts: 820
  • Karma: 47
    • Botanist, gardener, and preservice science teacher.
    • View Profile
  • Koppen zone: Dfb Googled
  • Hardiness Zone: USDA zone 6A
Re: Marker Assisted Selection
« Reply #3 on: 2021-04-06, 09:36:29 AM »
Conversation with the researchers making the offer makes the most sense to me.

I took one graduate non-degree course at OSU in plant genetics. Next steps towards modern plant breeding might be the skills to actually do marker assisted selection or direct gene selection. So if the offer is sans intellectual work me I would want a masters out of it. Though masters opportunities can be a way to get this sort of work done. If there is funding available someone could study these populations, and get a masters in the process or as a portion thereof. The intellectual work gets done and we get the information on which specific plants to cull and keep.

The link to the DNA snip looked like a habrochaites version of one of the important genes. I think it would make more sense to test for the domestic version. If the domestic version is entirely absent to my mind that would be the keeper plant.

The first article talks about a couple factors. Might be simpler to focus on just one of them initially- the one involved with that DNA snip in the last link.

We may have created complicating factors with our three species mixes and multiple wild accessions. However we may be able to ignore those and focus primarily on habrochaites and domestic versions of genes. The simpler cleaner population might be say a single habrochaites accession with one tomato variety such as Big Hill. However we don't have eight tomato generations to make the simpler version and it might not matter. Perhaps penellii genes will throw some weird results indicating if we needed to look at other possibilities. Or maybe absence of the domestic gene is adequate info.

The whole idea of doing the intellectual work around this has my brain wanting to bag one inflorescence on each plant to test for SI then plant mother to row.

What was the protocol last year in Idaho for determining SI?

Observation and record keeping alone?

The two resulting 2020 Idaho accessions shared with me R18 and S35S36S37 are labeled as R18 having better SI than S35-37 but both having SI. Is R18 the offspring of a single plant and S35-37 three plants combined?

In the XA1 / XA2 population in 2020 there was one plant that I thought was SC most I thought were SI.

The XL red population the first ripe I thought was SI based on blossom drop and then stopped keeping track but generally thought they were SI based on blossom drop. Though the consistency of the redness seemed consistent with SC.

That in mind XA1 / XA2 were mostly bicolor and that might indicate some SC. Though fruit consistency alone does not necessarily indicate SC could also suggest inbreeding with close relatives.

The best flavor of 2019 population mostly reverted to wild type or near wild type. I suspect they had good SI just because of that alone.
« Last Edit: 2021-04-06, 11:13:09 AM by William S. »
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

Joseph Lofthouse

  • Administrator
  • Full Member
  • *****
  • Posts: 342
  • Karma: 44
  • Great Basin desert, Rocky Mountains
    • Open Source Plant Breeding Forum, founder. World Tomato Society, ambassador. Plant Breeder. Yogi. Shaman.
    • View Profile
    • Garden.Lofthouse.com
    • Email
  • Koppen zone: Dsa
  • Hardiness Zone: USDA Zone 5
Re: Marker Assisted Selection
« Reply #4 on: 2021-04-06, 11:26:41 AM »
I reached out to a company that does marker assisted selection. One of the services they offer is whole genome sequencing. Now I'm daydreaming about selecting for many traits at once.

Large petals
exerted stigmas
functional HT allele
s-alleles
Determinate
Brix
Fruit size and color

I also reached out to my collaborators, that offered the funding grant, to open a dialog about how to get the project done this summer. I'm contemplating asking each of the major grow-out sites to send in 128 samples.

I don't know how many of those traits are known, but I bet a lot of them are...

The HT allele in domestic tomatoes is broken by having base-pairs out of order. It would be easy to observe by DNA sequencing regardless of pedigree.

In the F3 generation, which was done in a greenhouse in California, we tested for SI (by selfing) prior to making the crosses. The broken domestic HT allele is recessive, therefore phenotype observation alone can never eliminate if from the population, just reduce it's prevalence. Doing that type of work, extends the  required growing season, which is problematic because my growing season is already so short.
 
In the Idaho grow out, we used dropped blossoms, or fruit set without seeds, as an indication that the plants were acting self-incompatible. That is messy, cause what if a plant is self-incompatible, but highly fecund? That's why I called the Q-series "panamorous", they were dropping blossoms, but were also highly fruitful. What if they really are self-incompatible? I'd hate to eliminate them on the basis of being fecund. Being self-incompatible doesn't necessarily equal blossom drop.

R18 was one plant. It had the most promiscuous looking flowers. 

S35/S36/S37 were a sibling group of 3 plants, also with flowers looking very promiscuous. Flavor was the best out of 108 plants.

The XL population is such an outlier that I don't know what to do with it. If nothing else, it's flower structure offers high out-crossing rates compared to domestic tomatoes.

The flower below is from R18, compared to a domestic flower.


William S.

  • Full Member
  • *****
  • Posts: 820
  • Karma: 47
    • Botanist, gardener, and preservice science teacher.
    • View Profile
  • Koppen zone: Dfb Googled
  • Hardiness Zone: USDA zone 6A
Re: Marker Assisted Selection, Promiscuous Tomatoes
« Reply #5 on: 2021-04-06, 02:42:48 PM »
Not sure if I qualify as a major grow out site but this has me thinking I should plant another flat of the 2020 Bicolors out of A1 / A2 to make sure I have enough if needed. It would need to go take its chances in the greenhouse as the grow light station is fully utilized and partitioned between myself and my wife, but it is probably warm enough out there as my wife has a lot going on in there already and has been heating it at night. Eventually I plan to direct seed a row of them but that doesn't really need to happen till towards the end of April or even mid may (last frost is May 15th). Do I qualify as a major grow out site?

If I had to send in 128 samples from plants currently sprouted and growing I would probably have something like this to contribute:

15 XL from 2020 first fruit (It was delicious I recall thinking it had a smooth flavor despite being red and there was blossom drop). I can't believe I planted 15 of these in one cell.

4 ugly cracking bicolor monster from 2019 (had to be an interspecies hybrid I think)

3 Lofthouse habrochaites cytoplasm

11 LA2329 6 From G2 5 from G1 for a Habrochaites check

8 R18 (Planted all, but low germ)

22 S35,36,37 (Planted all but low germ)

55 XA1/A2 G2 Bicolors

5 Big Hill for a domestic check

5 Exserted Orange- think this is domestic but there is a lingering doubt.

Total should = 128

Also curious about possible alternate genes that could confound our observations. Like how Siletz and some other Oregon cultivars from that era have that weird trait that causes early fruits to form with zero seed. Also noticed some blossom drop last year in Big Hill and Exserted Orange due to weather conditions I think.

I've never bagged blossoms but it could be a useful tool both for checking SI and potentially just for using one plant in a pot for multiple controlled crosses.

When I took that Oregon State plant genetics course the professor told us about how he used marker assisted selection and doubled haploids to make an impossible barley. Basically the problem is that using standard methods we might grow out 1000 seedlings to get one ideal one. However if we can't genotype them we might select the wrong one. So you make 1000 doubled haploid barley plants and then use marker assisted selection to select that one seedling with the perfect combination of traits.

Also he explained that marker assisted selection is slowly being replaced with direct selection as they figure out exactly which sequences code for the traits. It has been nine years so probably fewer markers and more direct snippets of the actual trait are used now.

So seven traits sounds like exactly the promise of marker assisted or direct gene mapped selection. Just need enough individuals in the population with the potential for the traits to find a few perfect or near perfect ones. I suspect sometimes they cross two near perfect and repeat. Whole genome sequencing is another level. There are companies that will maintain a clonal tissue culture line of something, found one in Eastern Montana online once. Useful for pricey GMO's. Wonder if that could be a route to keeping genotyped elites for a few rounds of breeding.

The local seed potato growers send samples to Montana State University for disease checks. Any infected plants are weeded out. I sort of think this process could be similar. Gene check the population, then go through and destroy the ones indicated as faulty. Though it might actually be that we just only want to save seeds from the elites. Tomato plants produce a lot of seed.

Is one year, one round, of marker assisted (or better) selection adequate?
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

Joseph Lofthouse

  • Administrator
  • Full Member
  • *****
  • Posts: 342
  • Karma: 44
  • Great Basin desert, Rocky Mountains
    • Open Source Plant Breeding Forum, founder. World Tomato Society, ambassador. Plant Breeder. Yogi. Shaman.
    • View Profile
    • Garden.Lofthouse.com
    • Email
  • Koppen zone: Dsa
  • Hardiness Zone: USDA Zone 5
Re: Marker Assisted Selection, Promiscuous Tomatoes
« Reply #6 on: 2021-04-06, 03:21:50 PM »
Yes, William, I consider you to be one of the three major growers.

The literature that I was reading says that they can make a model, based on genome wide QTL traits. It would just be a matter of whether or not QLT loci are known for the traits that are of value. They are known for the HT and s alleles.

One round of marker assisted selection could stabilize the population into the wild allele of the HT gene.

QTLs for flavor would be a different matter.


William S.

  • Full Member
  • *****
  • Posts: 820
  • Karma: 47
    • Botanist, gardener, and preservice science teacher.
    • View Profile
  • Koppen zone: Dfb Googled
  • Hardiness Zone: USDA zone 6A
Re: Marker Assisted Selection, Promiscuous Tomatoes
« Reply #7 on: 2021-04-06, 03:39:15 PM »
Cool,

I just planted another 72 cell tray of the bicolor G2 XA1/A2 from the packet with the largest fruit and that fruit I Photographed with the black smudge on top- same trait I asked you about in 2019 and you said it probably came from one of the domestics. Looked nice on a bicolor. In the backyard greenhouse.
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

Andrew Barney

  • Full Member
  • *****
  • Posts: 472
  • Karma: 40
  • Northern Colorado, Semi-Arid Climate, USA
    • Pea Breeding, Watermelon x Citron-melon, Purple Foliage Corn, Wild Tomatoes
    • View Profile
    • My blog
    • Email
  • Koppen zone: Dfc / Dfb
  • Hardiness Zone: 5b
Re: Marker Assisted Selection, Promiscuous Tomatoes
« Reply #8 on: 2021-04-06, 10:20:57 PM »
Sounds like a very interesting offer. I think if you can swing it then don't let that opportunity pass by!

The trick is whether all the traits you care about have known markers. At this point most probably do.

What is the HT allele?

Are you selecting for the green shoulder gene in fruits? Even if not selecting for it directly it may have some linkage with higher brix.

I'll be doing my best this season to work on the known S. peruvianum hybrids that I got from the various gene banks. Last time I helped in any significant way I was focused on the S. pennellii hybrids and after that this project seemed to explode. If would be cool if in another few years the S. Peruvianum hybrids did the same effect like the pennellii genetics.

I personally think S. pennellii and S. peruvianum are the most interesting of the wild tomato species. The only ones after that are the very distant S. sitiens & S. lycopersicoides as they both have scented flowers. Would be awesome to restore scented flowers to tomatoes! But they are so distant I'm not sure if they would cross easily or not.

Joseph Lofthouse

  • Administrator
  • Full Member
  • *****
  • Posts: 342
  • Karma: 44
  • Great Basin desert, Rocky Mountains
    • Open Source Plant Breeding Forum, founder. World Tomato Society, ambassador. Plant Breeder. Yogi. Shaman.
    • View Profile
    • Garden.Lofthouse.com
    • Email
  • Koppen zone: Dsa
  • Hardiness Zone: USDA Zone 5
Re: Marker Assisted Selection, Promiscuous Tomatoes
« Reply #9 on: 2021-04-06, 10:38:38 PM »
I wonder if the scented flowers on S. sitiens & S. lycopersicoides are due to nectaries? The honeybees would love that.

Garrett Schantz

  • Full Member
  • *****
  • Posts: 223
  • Karma: 13
    • View Profile
    • Email
  • Koppen zone: Dfa
  • Hardiness Zone: USDA zone 6
Re: Marker Assisted Selection, Promiscuous Tomatoes
« Reply #10 on: 2021-04-06, 11:59:57 PM »


What is the HT allele?



The HT allele in domestic tomatoes is broken by having base-pairs out of order. It would be easy to observe by DNA sequencing regardless of pedigree.

In the F3 generation, which was done in a greenhouse in California, we tested for SI (by selfing) prior to making the crosses. The broken domestic HT allele is recessive, therefore phenotype observation alone can never eliminate if from the population, just reduce it's prevalence. Doing that type of work, extends the  required growing season, which is problematic because my growing season is already so short.
 
In the Idaho grow out, we used dropped blossoms, or fruit set without seeds, as an indication that the plants were acting self-incompatible. That is messy, cause what if a plant is self-incompatible, but highly fecund? That's why I called the Q-series "panamorous", they were dropping blossoms, but were also highly fruitful. What if they really are self-incompatible? I'd hate to eliminate them on the basis of being fecund. Being self-incompatible doesn't necessarily equal blossom drop.


https://pubmed.ncbi.nlm.nih.gov/11874575/

This should help with explain some things.

The defects in the HT / S-RNase alleles pretty much prevent pollination with S.peruvianum - some other species.

Usually we look for dropped flowers to indicate SI. But some of these tomatoes just drop flowers every now and again due to being highly exerted, fruitful, among other things.

SI plants require a separate plant as a pollen donor - this could include one of the panamorous types "thought" to be SI as a pollen donor. Ending up with pure SI hybrids is difficult.

Can't really phenotype non-broken HT alleles or SI plants very well. Also, as Joseph mentioned - the broken domestic HT allele is recessive - not so fun in a SI population. That is why marker-assisted selection is looking pretty nice.


If we manage to get Peruvianum hybrids without using embryo rescue, attempting crosses with S. sitiens  / S. lycopersicoides could be the next move. S. ochranthum / S. juglandifolium could be interesting as well.

S. peruvianum hybrids are created using embryo rescue - usually cuts off a bunch of genetic information from one of the parents. The broken HT allele issue is probably still there from the domestics - probably won't cross back to S. Peruvianum unless you get really lucky with the embryo rescue scramble. Though there are probably some fun genetics to play around with regardless.

One of the Peruvianums that I have is supposedly a SC accession - it is exerted though. I will try screening for SI domestic hybrids and trying for crosses with the SC Peruvianum. Of course I am isolating the SC Peruvianum from the other accessions - would rather not further complicate things. Probably won't work, but might as well try it out.

https://link.springer.com/article/10.1007/BF01253997

Hopefully a SC exerted Peruvianum will bypass the S-RNase problem. I would just need a domestic x wild that happens to be missing the HT allele issue.

If that all works, I could use the SC Peruvianum to introduce a bunch of other genetics from SI accessions.

I haven't grown the accession before, so I will have to confirm it is truly SC. If not I could try requesting SC germplasm next year.
« Last Edit: 2021-04-07, 12:17:49 AM by Garrett Schantz »

Joseph Lofthouse

  • Administrator
  • Full Member
  • *****
  • Posts: 342
  • Karma: 44
  • Great Basin desert, Rocky Mountains
    • Open Source Plant Breeding Forum, founder. World Tomato Society, ambassador. Plant Breeder. Yogi. Shaman.
    • View Profile
    • Garden.Lofthouse.com
    • Email
  • Koppen zone: Dsa
  • Hardiness Zone: USDA Zone 5
Re: Marker Assisted Selection, Promiscuous Tomatoes
« Reply #11 on: 2021-04-07, 12:50:52 AM »
I haven't done any selection for the green-shoulder trait on fruits. (Other than whatever selection is inadvertent because they taste great.)

I would say that the trait is present in the [Big Hill X Wild]-XL population, but not very prevalent in the general (Q-series) population.

William S.

  • Full Member
  • *****
  • Posts: 820
  • Karma: 47
    • Botanist, gardener, and preservice science teacher.
    • View Profile
  • Koppen zone: Dfb Googled
  • Hardiness Zone: USDA zone 6A
Re: Marker Assisted Selection, Promiscuous Tomatoes
« Reply #12 on: 2021-04-07, 03:19:09 PM »
Neat photos. S35, S36, and S37 remind me a lot visually of the XA1/A2 3-species seed packet outcomes. I need more taste testers. When are you three coming to visit this August/September?
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days

Andrew Barney

  • Full Member
  • *****
  • Posts: 472
  • Karma: 40
  • Northern Colorado, Semi-Arid Climate, USA
    • Pea Breeding, Watermelon x Citron-melon, Purple Foliage Corn, Wild Tomatoes
    • View Profile
    • My blog
    • Email
  • Koppen zone: Dfc / Dfb
  • Hardiness Zone: 5b
Re: Marker Assisted Selection, Promiscuous Tomatoes
« Reply #13 on: 2021-04-07, 03:31:29 PM »
I wonder if the scented flowers on S. sitiens & S. lycopersicoides are due to nectaries? The honeybees would love that.

Not sure,  but from what I remember reading one species had sweet smelling flowers and the other species had stinky flowers. Presumably each attracts different pollinators such as bees vs beetles/flies.

But,  yes I think that trait would be very interesting to honey bees. Could potentially increase fruit set if that were the case.

I have a few seeds from each species I will try to grow out this year. I found out from TGRC that they are genetically compatible and fully fertile with each other,  so who knows if they will cross as I grow them out. They are self incompatible species.

William S.

  • Full Member
  • *****
  • Posts: 820
  • Karma: 47
    • Botanist, gardener, and preservice science teacher.
    • View Profile
  • Koppen zone: Dfb Googled
  • Hardiness Zone: USDA zone 6A
Re: Marker Assisted Selection, Promiscuous Tomatoes
« Reply #14 on: 2021-04-07, 08:43:57 PM »
The notes about S36 intrigue me. It was the dwarf. I haven't noticed any signs of dwarf project type seedling dwarfism in the S35,S36,S37 tray but I will look closely again. Also I strongly suspect that there are multiple small size traits and the dwarf project only utilizes one of them. Edit: size variation in the S35, S36, S37 seedling tray is extreme. So my judgement is that the traits leading to small size in S36 are extant and currently expressed in the population.

In general as an observation based on my trays tomato seedling size does seem to have a strong genetic component from an early age. I am curious to know if the determinant gene is also indicated by short seedling stature. Indeterminate Brad is one of the taller accessions. Plants closer to the light are also shorter. More even lighting would be necessary for better comparisons on different parts of trays.
« Last Edit: 2021-04-07, 09:12:13 PM by William S. »
Western Montana garden, glacial lake Missoula sediment lacustrian parent material and shallow 7" silty clay loam mollisoil topsoil sometimes with added sand in places. Zone 6A with 100 to 130 frost free days