From the lab - family tree of White Labs yeast

[ba-brewer]

Thanks for the dot info and posting the video. In the middle of the video they showed the strain chart with POF and other characteristics identified or quantified. Is that available somewhere on the web?

I slanted wlp030 and a few others when they were released from the vault. I like the idea of the vault but something needs to change to get strains released a bit faster, maybe get breweries or suppliers to sponsor strains.

 

[Northern_Brewer]

They say it's from Gallone et al which was the original Leuven/White Labs sequencing paper, but I can't see it from a quick scan, it may be that Brewlab have done their own figure based on eg fig 5 in that paper.

Something that would make a significant difference, certainly for British strains, would be opening up the Vault to non-US customers - at the moment only people with US credit cards can express an interest. It also seems badly arranged how the Vault ignores seasonal releases - but at the moment White Labs seem in such chaos that any changes would probably just make things worse.

 

[Mer-man]

Yeah, I mean there's a white labs facility here in Copenhagen.

 

[kmarkstevens]

It would be nice to compare WLP002 to 1968 to a sample of yeast from a Fuller's cask or bottle to see which of the commercial strains is a match. It seems like neither of them might be the real deal.

Pub supposedly is also the same yeast, but I liked it better in side by side comparisons. Even using only first gold hops, I couldn't get a hint of marmalade. I think you may be correct this isn't the fuller strain

 

[kmarkstevens]

As an aside, at the weekend I had it confirmed from John Keeling himself that it's the production yeast in the bottle - they centrifuge all their bottled beer clean, then reseed with 0.5 million cells/ml. Although they only use the "single" strain they've used since 1976 in everything now, even the historical recreations, apparently it now looks like two strains on DNA.

Does this imply I could culture the Fuller's strain from London Pride?

 

[Northern_Brewer]

Does this imply I could culture the Fuller's strain from London Pride?

Sorry, I should have been more explicit, the above only applies to their bottle-conditioned beers. In the UK that means 1845 and Bengal Lancer of their core range, and specials like Vintage. But Pride, ESB etc are pasteurised and not bottle-conditioned, so they have no yeast.

 

[kmarkstevens]

@Northern brewer. Thanks, I guessed that would be the answer but was hoping that London Pride would work.

 

[Northern_Brewer]

This article is about the realisation a few years ago that "Brett" Trois wasn't actually Brett. I've not seen anything about how it fits into the main Saccharomyces family tree though. But it ends with a little teaser from Troels Prahl of White Labs Copenhagen :
"We’re coming out with a pretty interesting paper here about hybrids," Prahl says. "One of the things that was remarkable when we got all that data about the sequencing was how many strains around the world are actually hybrids—which means that there are going to be several new reclassifications."

Heard anything on the grapevine @suregork?

 

[kmarkstevens]

Calling all Yeastie Boys. White Labs is going to put all vault yeast into production on 1 July. Here's your chance to get something like WLP85 that is collecting an order every 6 months.

 

[kmarkstevens]

https://www.whitelabs.com/yeast-vau...244964713&mc_cid=4485d2e0c7&mc_eid=92ebff3dd0

 

[ba-brewer]

@suregork

Are there strains in the vault that you(or colleagues) would like get your hands on for testing? Is there a way people can make donation to help with your research? I would be willing to make an donation or procure the yeast and send them.

 

[David17]

This article is about the realisation a few years ago that "Brett" Trois wasn't actually Brett. I've not seen anything about how it fits into the main Saccharomyces family tree though. But it ends with a little teaser from Troels Prahl of White Labs Copenhagen :
"We’re coming out with a pretty interesting paper here about hybrids," Prahl says. "One of the things that was remarkable when we got all that data about the sequencing was how many strains around the world are actually hybrids—which means that there are going to be several new reclassifications."

Heard anything on the grapevine @suregork?

When this occurred, I was secondary fermenting a saison with Brett Trois. It was a surprise that a lab with WLP's history failed to know they had it wrong since they do the science on which we all depend. Of course, I use their products. I may yet have BT in my freezer .

Keep brewing - it's quite an adventure.
Keep drinking - and that's quite an adventure.
Live your live freely - the ultimate adventure.
Finally, Praise God for all you have.
David

 

[kmarkstevens]

I ordered WLP 11, 17, 26 and 85 from the upcoming Vault release to compliment the yeast library sitting in my fridge.

011 European Ale because it and 02 are the lowest attenuating yeasts out there and work well with low ABV beers
017 Whitbread II because it is at least half of the Whitbread strain
026 Premium Bittler because it is estery
085 English Ale yeast blend because I really like this one. More to it and a bit less attenuative than 02. It's a mix of 02 and 07 (?). Flocculates on a dime and gives you change. I picked it up at White Labs San Diego when I had about 15 minutes to do a flight of swifties and head over to see some relatives. It's been at about 2 orders for the past 18 months, so I want to have a back up since I'm on gen 4 or 5 at this point

 

[kmarkstevens]

Got an email from White Labs saying that today is the last day to get your vault order in.

they also added this teaser: After the strains are in production (and on their way to your doorstep) we're making updates to the Vault for Homebrewers. Look for easier ordering and more unique yeast strains to be released. Stay tuned for our announcement for more details!

 

[deadwolfbones]

I got:

WLP611 New Nordic Yeast Blend
WLP518 Opshaug Farmhouse Yeast
WLP665 Flemish Ale Yeast Blend

Excited!

Also my Leeuwenhoek Saison Blend came today.

 

[ba-brewer]

this quote was posted in the whitelabs vault thread, but the question may help other more in this thread
https://www.homebrewtalk.com/forum/threads/white-labs-yeast-vault.565113/page-9#post-8625513

You are really close to your mother on your family tree, but hopefully you are fairly different! So proximity doesn't always mean close similarity, particularly for single/few gene characteristics like phenol production or floccing, whereas they can be quite similar for multigene characteristics (like general esteriness or human height).

And actually the family tree in that area is not that close - there's few strains sampled, which means they are "close" as you go sideways around the circle but what matters is the "height" of the lines, and you'll see that there's quite "mountain" between 025 and 009, on that measure they're further apart than eg Nottingham/039 and the Whitbreads.

With respect to your comments been wonder what the average homebrewer can glean from the family tree if two strains can be closely related yet can product quite different results?

I have been looking at the chart and viewing the dots as all one point like a electrical diagram, but I am now thinking that might be the case. In the picture below is the highest dot a yeast that create WLP025 and WLP008 or is there another intermediate yeast(next dot down) between them?

Also could you provide a key to how to read the chart?

 

[dmtaylor]

is the highest dot a yeast that create WLP025 and WLP008 or is there another intermediate yeast(next dot down) between them?

The way I read this is a combination of the dots and the lines. After a dot, the longer the line is, the longer it took for the next two offshoots to develop. So, after that highest dot, there is some distance before the next offshoots, and it happens quicker on the left branch than the right branch, which means the left branch has fewer ancestors in between the original and the new branch than the one on the right. Then from there...

WLP025 for example took many many generations before it developed, given the very long line on that one.

Conversely, once that central branch developed, WLP009, beer095, beer097, and beer096 all happened very quickly with just a few ancestors in between each one so they are all very closely related. Not sisters but reasonably close to each other.

And then on beer042 and CSB1248, while they shared a common ancestor, by their long lines it looks like the ancestor they shared was sort of like a great-great-great-great-great-...-great-grandmother. Way way back... a century or two ago. So they are actually quite distant relations. But still family in the same way that a 7th cousin is related to you --- not very closely related actually!

 

[Northern_Brewer]

With respect to your comments been wonder what the average homebrewer can glean from the family tree if two strains can be closely related yet can product quite different results?

There is information to be had, but it's fuzzy and a bit complex, just in the same way that relationships within human families can be complex. I'm sure we can all think of families where the children come out almost identical, and others where they are very, very different.

Things like POF phenolics make a big difference, but are controlled by fairly simple switches. So Windsor and S-33 are pretty closely related and behave pretty similarly, but adding the POF "gene" takes you to T-58 which aside from the peppery phenolics actually behaves pretty similarly to S-33. But all people will tend to notice is the phenol character. But for instance all the Group 2 yeasts have a certain distinct character, and it's certainly helpful to understand that eg WB-06 is a Group 2 and nothing to do with the main hefe strains, which are effectively a subgroup of kolsch yeasts that have acquired a little fragment of Group 2 DNA.

I have been looking at the chart and viewing the dots as all one point like a electrical diagram, but I am now thinking that might be the case. In the picture below is the highest dot a yeast that create WLP025 and WLP008 or is there another intermediate yeast(next dot down) between them?

It's an attempt to guess an actual family tree, so the "junctions" represent the last common ancestor of a pair of strains. The junctions with a dot are ones with >95% probability of being right, emphasising that it's a guess. The key to realise is that the height of the lines represents genetic distance, which as a first approximation very crudely corresponds to time between two yeasts splitting off from a common ancestor until their evolution was "stopped" by being banked in a freezer (typically 20-30 years ago). The correlation between genetic distance and actual time is frustratingly imprecise and will be different for different strains as eg they will mutate faster in the stressful environment of a high-ABV Trappist beer, and will mutate slower if they're only being used to ferment once a year like wine yeast and saisons. Sure enough, there's a lot less variation between wine yeasts - the critical factor is the "height" of the lines, sideways distance is purely cosmetic depending on how many strains have been sequenced in a particular group. But the Leuven lab claim that Group 1 and 2 split off about 400 years ago so that gives you a rough idea of the timescales involved, although I suspect it's a bit older than that.

Apologies to all the biologists here, the above is just trying to give a flavour of what's involved without writing an essay!

 

[ba-brewer]

Thanks for the explainations @Northern_Brewer and @dmtaylor.

 

[kmarkstevens]

I got:

WLP611 New Nordic Yeast Blend
WLP518 Opshaug Farmhouse Yeast
WLP665 Flemish Ale Yeast Blend

Excited!

Also my Leeuwenhoek Saison Blend came today.

Got an update today the Opshaug is released from the vault. My past experience means it's about 3 weeks from my door.

I previously got the 611 New Nordic Yeast Blend. did do much for me on the cyser front, so hope you have a better exerperience

 

[Northern_Brewer]

There's a new preprint out from the Dunham lab in Seattle that is mostly looking at what happens to yeast genomes as they are repitched, but they also sequenced all the Chicos they could find to get an idea of the background diversity.

The sequences were filtered against Gallone BE051 (almost certainly WLP008 East Coast) as the most divergent US strain they had.

You can see two main groups in the bottom of the chart (Fig 2 in the paper, which I've annotated in green), one based around Wyeast 1056 that includes US-05 and Imperial A07 Flagship (and "Imperial M44" - they're double checking that, it may be Mangrove Jack M44??), and one group based around White Labs WLP001 that includes the Chicos from Gigayeast, Escarpment, Craft Cultures, Omega and the old Wyeast VSS strain 1792 (Fat Tire).

If I read it right, I think they're saying that a hypothetical ancestor (at Sierra Nevada ?) of the 1056 and WLP001 groups had a A234D mutation in BAT1, a branched-chain amino acid(BCAA) aminotransferase, but the WLP001 group has a recombination on one end of chromosome VIII that restores the wild type. This seems to be a weak point in the chromosome, as the recombination seems to happen a fair bit independently. They're submitting another paper just on the BAT1 story, but "the mutation led to a number of phenotypes including a sensitivity to osmotic stress, reduced fermentation ability when grown in 20% glucose [~1.075], and a growth defect in minimal media" Per Table 3 the mutants also produce more isoamylacetate ("banana") and fusel alcohols. However the strains with the chromosome recombination don't ferment out quite as fully, and also show more diacetyl.

Both the WLP001 and 1056 groups mostly have just three copies of chromosome V, but their putative ancestor BRY-96 has four copies (green and blue squares respectively in diagram). Again this seems to be something that is rather fluid, but they observed four copies arising in a population of 1056 and spreading as it was repitched in a brewery. It could be something that gives a selective advantage in some environments but not others, or it could be just a random thing that happened at a bottleneck on the way to 1056 and WLP001.

I read "The strain BRY-96, which is used in Elysian Brewing Co." as meaning that Elysian buy in BRY-96 direct from Siebel/Lallemand, their version is certainly very close to a version sourced from Lallemand. The tree fits previous suggestions that A30 Corporate is derived from Elysian one way or another and it's the closest strain to BRY-96 that's commercially available, but the strain sourced from Elysian is closer to BRY-96 than it is to A30. Also it looks like Bell's yeast is a BRY-96 that's spent some time evolving away from Siebel.

It's interesting the way that the "San Diego/Pacman" group are distinct from the main BRY-96 family - and how did BE082 get to Belgium? @suregork and friends were already pretty certain that BE071 was WLP090 and this rather confirms it (also BE098 as 1056, and this tree rather suggests BE102 was US-05)

However, one surprise is that they reckon that NRRL Y-7408, isolated from Ballantine in 1972, is probably a cousin rather than an ancestor of the BRY-96 group, based on "large segments of variation" that it has lost compared to the BRY-96 group. Presumably BE065 has also lost them.

BE065/68/69 were the main uncertainties in the main run of White Labs yeasts in the original Gallone paper. Two obvious candidates are
WLP019 California IV and WLP076 Old Sonoma (New Albion) but we can't be sure - BE044-086 were pretty much all the White Lab strains in numerical order, but they may have eg had some strains from commercial customers of theirs in there somehow. Is SNPA BE007 or BE068?

There's also a ton of detail on how yeast genomes change over time, but that's another story....

 

[kmarkstevens]

Interesting. I'm going to have to read that when I have more time.

FYI PostDoc Brewing is just around the corner from my LHBS, which is next to Mac n Jacks brewery. The Post Doc dude is a University of Washington graduate, which is where Maitreya Dunham does her research. She had a previous project with pichia apotheca, and very nicely sent a sample on request.

 

[Wolfbrau]

The big confounding factor, insofar as tracing a lineage is concerned, seems to be that essentially all breweries used multistrains up until shockingly recently, while yeast banks have been using single strains for virtually the entirety of their existence.

 

[VikeMan]

Interesting. I currently have a Fat Tire clone fermenting with US-05, mainly because the lastest @suregork tree shows US-05 to be close to the not available Wyeast 1792 (the purported New Belgium house strain). The Suregork tree also shows US-05 and Wyeast 1792 to be closer to each other than either is to WLP001.

But the Dunham Lab tree above puts WLP001 very close to Wyeast 1792, with US-05 off in the relative hinterlands.

 

[suregork]

Interesting. I currently have a Fat Tire clone fermenting with US-05, mainly because the lastest @suregork tree shows US-05 to be close to the not available Wyeast 1792 (the purported New Belgium house strain). The Suregork tree also shows US-05 and Wyeast 1792 to be closer to each other than either is to WLP001.

But the Dunham Lab tree above puts WLP001 very close to Wyeast 1792, with US-05 off in the relative hinterlands.

The Dunham Lab tree, looking at only strains in a specific group and with high quality sequence data, is definitely more accurate than the my big tree!

 

[Northern_Brewer]

There's a new preprint out from the Dunham lab in Seattle that is mostly looking at what happens to yeast genomes as they are repitched, but they also sequenced all the Chicos they could find to get an idea of the background diversity.
View attachment 687412

If I read it right, I think they're saying that a hypothetical ancestor (at Sierra Nevada ?) of the 1056 and WLP001 groups had a A234D mutation in BAT1, a branched-chain amino acid(BCAA) aminotransferase, but the WLP001 group has a recombination on one end of chromosome VIII that restores the wild type. This seems to be a weak point in the chromosome, as the recombination seems to happen a fair bit independently. They're submitting another paper just on the BAT1 story, but "the mutation led to a number of phenotypes including a sensitivity to osmotic stress, reduced fermentation ability when grown in 20% glucose [~1.075], and a growth defect in minimal media" Per Table 3 the mutants also produce more isoamylacetate ("banana") and fusel alcohols. However the strains with the chromosome recombination don't ferment out quite as fully, and also show more diacetyl.

I missed the BAT1 paper a few weeks ago, it's here : Effect of the Ala234Asp replacement in mitochondrial branched-chain amino acid aminotransferase on the production of BCAAs and fusel alcohols in yeast (paywalled, I've only read the abstract, I don't suppose someone has access? )

So they've done some genetics on it and it seems the A234D mutation has the same effect as deleting BAT1 entirely, even though it "shouldn't" affect it that badly based on homology studies. In both cases the mutation (ie BRY-96/Corporate/Bell's and the 1056/US-05 group, but not the WLP001 group) reduces growth in the log phase, reduces intracellular valine and leucine, increases fusels and reduces CO2 production.

 

[Andre3000]

There's a new preprint out from the Dunham lab in Seattle that is mostly looking at what happens to yeast genomes as they are repitched, but they also sequenced all the Chicos they could find to get an idea of the background diversity.
View attachment 687412

The sequences were filtered against Gallone BE051 (almost certainly WLP008 East Coast) as the most divergent US strain they had.

You can see two main groups in the bottom of the chart (Fig 2 in the paper, which I've annotated in green), one based around Wyeast 1056 that includes US-05 and Imperial A07 Flagship (and "Imperial M44" - they're double checking that, it may be Mangrove Jack M44??), and one group based around White Labs WLP001 that includes the Chicos from Gigayeast, Escarpment, Craft Cultures, Omega and the old Wyeast VSS strain 1792 (Fat Tire).

If I read it right, I think they're saying that a hypothetical ancestor (at Sierra Nevada ?) of the 1056 and WLP001 groups had a A234D mutation in BAT1, a branched-chain amino acid(BCAA) aminotransferase, but the WLP001 group has a recombination on one end of chromosome VIII that restores the wild type. This seems to be a weak point in the chromosome, as the recombination seems to happen a fair bit independently. They're submitting another paper just on the BAT1 story, but "the mutation led to a number of phenotypes including a sensitivity to osmotic stress, reduced fermentation ability when grown in 20% glucose [~1.075], and a growth defect in minimal media" Per Table 3 the mutants also produce more isoamylacetate ("banana") and fusel alcohols. However the strains with the chromosome recombination don't ferment out quite as fully, and also show more diacetyl.

Both the WLP001 and 1056 groups mostly have just three copies of chromosome V, but their putative ancestor BRY-96 has four copies (green and blue squares respectively in diagram). Again this seems to be something that is rather fluid, but they observed four copies arising in a population of 1056 and spreading as it was repitched in a brewery. It could be something that gives a selective advantage in some environments but not others, or it could be just a random thing that happened at a bottleneck on the way to 1056 and WLP001.

I read "The strain BRY-96, which is used in Elysian Brewing Co." as meaning that Elysian buy in BRY-96 direct from Siebel/Lallemand, their version is certainly very close to a version sourced from Lallemand. The tree fits previous suggestions that A30 Corporate is derived from Elysian one way or another and it's the closest strain to BRY-96 that's commercially available, but the strain sourced from Elysian is closer to BRY-96 than it is to A30. Also it looks like Bell's yeast is a BRY-96 that's spent some time evolving away from Siebel.

It's interesting the way that the "San Diego/Pacman" group are distinct from the main BRY-96 family - and how did BE082 get to Belgium? @suregork and friends were already pretty certain that BE071 was WLP090 and this rather confirms it (also BE098 as 1056, and this tree rather suggests BE102 was US-05)

However, one surprise is that they reckon that NRRL Y-7408, isolated from Ballantine in 1972, is probably a cousin rather than an ancestor of the BRY-96 group, based on "large segments of variation" that it has lost compared to the BRY-96 group. Presumably BE065 has also lost them.

BE065/68/69 were the main uncertainties in the main run of White Labs yeasts in the original Gallone paper. Two obvious candidates are
WLP019 California IV and WLP076 Old Sonoma (New Albion) but we can't be sure - BE044-086 were pretty much all the White Lab strains in numerical order, but they may have eg had some strains from commercial customers of theirs in there somehow. Is SNPA BE007 or BE068?

There's also a ton of detail on how yeast genomes change over time, but that's another story....

I must have missed this post! Some great stuff here. I take it my best alternative if I want to get close to Elysian / A30 / BRY-96 group might be Bells? A guy can order it from the Bells store. I keep wondering if there's a more common WLP equivalent in there somewhere.

 

[Northern_Brewer]

I must have missed this post! Some great stuff here. I take it my best alternative if I want to get close to Elysian / A30 / BRY-96 group might be Bells?

In short - yes.

Although Lallemand BRY-97 may also work, although nothing's been published it seems like it's an early member of the group, but might have the BAT1 mutation that would put it as an ancestor of the 1056/US-05 subgroup.

And usual caveat that phenotypes can differ even between quite close relatives, as you can see between any mother and son.

 

[Northern_Brewer]

I grabbed this from a recent Fermentis seminar, it's the first time I've seen any kind of family tree from Fermentis. Unfortunately it's not great quality and doesn't have many comparators, which reduces the potential precision. Also they've done it the "wrong way round" compared to the ones we're used to, so I've flipped and rotated it.

I guess the biggest "surprise" is BE-256 showing up in the middle of the English yeasts, which immediately reminds one of WLP540.

Also K-97 looks suspiciously close to being in what I assume are the lager yeasts, although you'd expect a kolsch strain to be close to that region.

Would be interesting to know what wine yeast F-2 is, and what they're using in what I think is HA-18 (it's the same photo as in their catalogue, but I can't be sure that's what it is).

Pinging @dmtaylor and @suregork if they've not seen this before.

 

[dmtaylor]

@Northern_Brewer , no I have not seen this before now. Thanks! I will study it in detail tonight or tomorrow. Awesome.

EDIT: 2 hours later.....

I found some time.

BE-256 is indeed interesting. While I doubt it is very closely related to WLP540, I can agree it looks like it could share English heritage in that wing of the family tree. It might be one of those weird interesting mutants that manages to attenuate high and produce a good bit of isoamyl acetate (banana) -- I'm thinking along the lines of the POF+ English yeasts, except on this one it's POF- but full of banana. A mutant like this seems rare, but it could happen I suppose. That, or their study is just erroneous on this one. Almost a little too surprising to be very believable.

K-97 was already listed next to 1007 on @suregork 's page, and while relatively clean and similar to 1007 in its fermentation characteristics, it's not what I would consider to be particularly lager-like in my experience, so I'm not sure I would trust a leap toward the lager family. Or possibly, it's some sort of blend of alt and lager yeast strains so that it could show up in two places. I know personally I'm not happy with K-97 in my own brewery, and won't use it again. Wyeast 1007 turns out better IMO -- this one would be my go-to altbier yeast in future. I am not aware of any White Labs "equivalent". Bottom line here, I think K-97 stays over by 1007.

The other one I find interesting is BE-134. In my own table, I had placed this one next to WLP590 "French Saison". I don't recall my basis, I'd have to dig and ponder some more if needed. But now, based on the above image, I'm wondering if BE-134 could be more akin to like WLP351, the so-called "Bavarian Weizen" strain that might actually be pastorianus!? With all of these being POF+ and high attenuation, it's hard to figure exactly what fits where, but what's clear is that they are all in that same mosaic neighborhood. That at least makes sense.

The others all appear consistent with what we knew previously. I myself remain ignorant of the wine strains and bottling strains, as my focus is really just beer yeast, but if anyone wants to direct me where I should stick them on my table, I suppose I would consider it.