Mapping yeast mutations?

[Eigenbrau]

Do we have any biologists here?

I've been doing some dabbling in biohacking and was curious whether or not one would be able to identify yeast mutations using a gel electrophoresis. My theory was that it may be possible to create a reference map once you receive your initial yeast specimen, then every other generation or so, remap it using the gel and compare to the original.

The only issue I could see with this is just how much would the genetic makeup change? Would it be visible? Even if it does change, how would one go about identifying the mutation and ensuring it's not a bad one?

I've refused to take "you can't do that" as an answer when it comes to identifying mutations

 

[adc123]

I have thought of something along these lines too. My first worry is having enough DNA for a test. Turns out you don't need that much. I found this very useful:

http://www.bioteach.ubc.ca/TeachingResources/DoingScience/MacgyverProjShirazuEtalMaintext.pdf

Describes how to prepare DNA for gels. Other than the equipment hurdle, its probable this can happen. I just wonder how it would well it would work every time. I'll ask my bio friends about this!

The better question is, can we map the gels to real changes in the yeast. I talked to a craft brew guy that said he reused some us-05 MANY times with no ill effects. It's quite possible the yeast doesn't change but you harbor small amounts of bacteria with each batch, which eventually catches up with your taste buds.

 

[adc123]

The setup can also be cumbersome. DNA sequencing is really the way to go here. I found one company that might do them one at a time:

http://www.functionalbio.com/index.htm

You may or may not have to provide them just DNA; looks like they may be able to extract it from the yeast. Time to get super nerdy...

 

[chicagobrew]

Send a PM to peoplesbrewingcoop. He's currently sequencing yeast DNA and could answer a lot of your questions.

 

[Eigenbrau]

I asked a couple of the microbiologists on Reddit and they've stated it'd probably be very difficult, if not completely impossible to do in your garage. While electrophoresis is easy and cheap to run in your garage, they said the resolution of the gel would not be enough to pinpoint individual gene mutations. You'd also have to know which restriction enzymes to use to break the yeast's DNA up.

I'm curious to see a microbiologist get hold of this in the garage and give it a shot though!

 

[adc123]

At least for a next few years, it will be hard, but with the amount of biotech starting up, it will become easier to sequence with "homebrew" electronics. I would find it interesting to send a few samples to the company I listed and just compare Wyeast 1056 to US-05. But thats just me.

 

[maddprofessor]

You would most likely have to do a PCR and then sequence it as mutations are usually a change in the DNA which results in a change in the protein from one amino acid to another... which doesn't change the size of the DNA so it would look the same on the gel. If you cut the DNA with restriction enzymes you would only see a difference if the mutation happened to occur in a very specific spot. I currently am working on my PhD and mapping mutations in hepatitis C virus and it's small genome is enough of a chore to work with.

 

[Eigenbrau]

I just found a good Powerpoint on pulsed field gel electrophoresis (PFGE). It MIGHT be possible to do this in the garage with the correct restriction enzymes and some knowledge of common mutation areas...I'll do my own research when I get home in a few weeks and keep you guys updated.

 

[broadbill]

I just found a good Powerpoint on pulsed field gel electrophoresis (PFGE). It MIGHT be possible to do this in the garage with the correct restriction enzymes and some knowledge of common mutation areas...I'll do my own research when I get home in a few weeks and keep you guys updated.

Its not so much that the techniques are outside the scope of what can be done in a garage/lab...it is the project itself which is cumbersome. The yeast genome is 12 million base pairs in length. That's alot of restriction cutting and gel-running to find a mutation somewhere in a genome of that size.

Next, you are assuming any phenotypic changes observed in yeast behavior (alcohol tolerance, attenuation) would be a direct result of a gene mutation. There are epigenetic changes to consider that might change a yeast strain's phenotype without changing its genotype.

You are also assuming that you would be able to detect mutations should they exist just by restriction cutting and gel electrophoresis. Keep in mind that you would be searching for mutations in a mixed population of yeast strains (mutated vs. non-mutated is the simplest example). Would you be able to see a difference on a gel if your yeast population consisted of 20% mutated vs. 80% non-mutated? Unlikely. Furthermore, your mixed population will most likely contain yeast cells with many different mutations as well as yeast that have multiple, different mutations within their genomes. It would get complicated quick.

Just a few hurdles! Welcome to lab work...

 

[adc123]

I am no stranger to the proper design of experiment either

Thanks for the input and new words to google!

 

[Eigenbrau]

It's good to have someone with the insight/education to objectively look this over. I found the link for the Powerpoint I have saved should anyone wish to look it over:

http://www.mbaa.com/Districts/milwaukee/ppts/YeastFingerprinting.ppt

It had some good information, but is lacking in a lot of aspects, and I think the hurdles/roadblocks you've put in place would definitely put this beyond the scope of my patience and more importantly, my budget.

 

[Naked_Eskimo]

Having done plenty of PFGE in my lifetime, you can pretty much forget doing this "in the garage". Besides the expense of the equipment and reagents, it's not a trivial procedure to do correctly.

While PFGE is great for separating large pieces of chromosomal or plasmid DNA, it's essentially impractical for identifying small scale mutations ... unless that mutation happened to occur in the very restriction enzyme site (probability 0.000000000001%). When you're separating out 100KB fragments, you're not going to notice a change in 1 or 10 or even 1000 bases.

Not to be a naysayer...but unless you're mister moneybags, this isnt going to happen.

 

[pjj2ba]

Your best bet would be to mutagenize a yeast strain using some kind of transposable element or some recoverable marker. Then comes the fun part. You take say 1000 independent lines (20 K would be best) from the mutagenized strain, make up a large batch of wort and then divide it up among 1000 fermenters, and let the yeast have at it. Bottle and taste. If you find one you like (and kept good records) you can go back the the line that was used for that batch. Since use mutagenized with a marker of known DNA sequence, you can use that to do PCR and recover the sequence flanking the insertion which most likely will contain the gene responsible for the superior brew.

 

[Anubis]

^ what he said.

 

[Eigenbrau]

Your best bet would be to mutagenize a yeast strain using some kind of transposable element or some recoverable marker. Then comes the fun part. You take say 1000 independent lines (20 K would be best) from the mutagenized strain, make up a large batch of wort and then divide it up among 1000 fermenters, and let the yeast have at it. Bottle and taste. If you find one you like (and kept good records) you can go back the the line that was used for that batch. Since use mutagenized with a marker of known DNA sequence, you can use that to do PCR and recover the sequence flanking the insertion which most likely will contain the gene responsible for the superior brew.

Damn, and I only have 998 fermenters

 

[avibayer]

I'm an undergrad getting my degree in biology, and i happen to be writing my term paper for genetics on the MAL gene (allows the yeast to absorb maltose, and then begin to break it down).

AngryMage, the project of identifying yeast mutations and genes that lead to a superior brew, is one that is heavily taken up by the yeast labs at certain breweries. There is a lot of literature available (though i only have access because of school subscriptions) out there.

Running gels is pretty easy, but wont give you the answers your looking for. check out what research is being down, and i am happy to send you any articles you might want to read. Also, a cool book, "froth", discusses population dynamics for pitching and the resulting temperature flux.

 

[The_General]

I'm repeating what everyone else is saying but the best/only way to do this is to grow up a clonal population and sequence it.

If you know EXACTLY what you're looking for and you're a little clever, you could find a single mutation by using a gel (or maybe a gel and PCR).

Even if you had the means to do this, you're dealing with a ton of data and would probably need an expert to sift through your data to figure out the differences and not all of them may mean anything.

This is "needle in a haystack" sort of stuff and even if you find mutants, you'd have a hard time figuring out what they mean.

 

[JohnMc]

, they said the resolution of the gel would not be enough to pinpoint individual gene mutations. You'd also have to know which restriction enzymes to use to break the yeast's DNA up.

That's right, regular gels don't have the resolution for most mutations, pulse field even less so.

As far as sequencing is going, there's an anticipated "$100 Genome", that's a hundred bucks for a person's entire DNA, with a hoped for turnaround time of less than a day. Look up Pacific Biosciences and also "next gen" sequencing.
As was said earlier, the yeast genome is about 1.25x10^7 base pairs, the human one is roughly 3.3x10^9, or about 9000 times bigger. Maybe we can get a discount on a small genome!

Another thing to read about if you've got the bug is association mapping, specifically genome wide association mapping. You take a whole bunch of individuals with different mutations and such and check them for traits (phenotypes) of interest and correlate the phenotype and genetic markers with the traits with very high resolution. It works best if you have a corresponding high resolution set of markers, which genome sequencing can give. The ~3-4 year old state of the art in this is SFPs.

avibayer, you focusing on one MAL gene or the whole set?

 

[avibayer]

Its a Genetics survey class, and the professor wants only one locus. so, i will mention all of them, but i think with focus on MAL6. Not too sure yet. I plan to go into all three genes on the locus though. Might go into detail about the regulation.

What i was amazed about was how many different strains of S. cerivisiae there are. Another reason i think mapping the genome of a specific batch of yeast, might be a later step in a bigger process.

here is an homage to the father of yeast genetics http://en.wikipedia.org/wiki/Øjvind_Winge

 

[JohnMc]

here is an homage to the father of yeast genetics http://en.wikipedia.org/wiki/Øjvind_Winge

Here's a good book to read, if you're inclined to read about the history of yeast genetics.
The Early Days of Yeast Genetics
Odds are, your school's library has a copy.

 

[avibayer]

They dont have it, but i'm having it sent over from another library. thanks for the recommendation.

 

[JohnMc]

They dont have it, but i'm having it sent over from another library. thanks for the recommendation.

You're welcome. I enjoyed it when I read it mumble, mumble years ago.

 

[ni*]

I do molecular genetics research on yeast. I'm writing to agree with everyone else: I have a very broad view of 'possible at home', but this basically isn't possible at home.

A few remarks, however. Gel electrophoresis isn't the right tool for this job -- no one would ever use gel electrophoresis to do this; there are vastly easier ways. (As others have mentioned, it wouldn't really work anyway, even if combined with some huge library of restriction enzymes. You'd miss almost all mutations. Of course, you'd die of old age long before.)

Having small segments of DNA sequenced isn't actually that expensive (it's completely routine in any molecular bio lab), and that's a very easy way to identify what mutation you're looking at, if you know where it is. This is definitely in garage price-range.

There's a much broader problem, though: For the purposes of brewing, the sequence encoding the mutation doesn't matter in the least. There is no reason I can think of why you would have any reason to care exactly what the sequence ended up being.

S. cerevisiae is an extremely convenient organism to do genetics on, not least because other people have been doing the hard work for us for the past fifty years. If you're willing to devote a fair bit of time, and a small but reasonable amount of money to it, you can easily do classical genetics on yeast in your garage. You won't need any equipment beyond a microscope. It's not trendy, and doesn't sound as cool as directly manipulating DNA, but classical genetics is entirely powerful enough to create and map mutations.

As for your original question regarding how much mutation you can expect, the answer (as you probably realize now) is "very little". If you want mutations, add a mutagen -- UV is probably the most convenient, although they'll all give you cancer if you're not careful.

If you have other questions, I can probably answer them, but a book on genetics would probably be your best bet.

 

[avibayer]

Ni, any suggestions for a place to find journal articles on MALx genes? My school has a ton of subscriptions. Being in the business, thought you might be able to point an aspiring researcher in the right direction.

 

[ni*]

Honestly, unless I know what I'm going for, I usually just use Google Scholar, which, in this case, turns up:
http://www.pnas.org/content/81/9/2811.short

You generally start with review articles (like that one) in this situation, and look at their references for anything that seems interesting. In this case, that review is pretty dated, so it can be helpful to do the reverse: look toward the bottom of the page at articles citing it.

http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2958.1991.tb02136.x/abstract (and related articles) is probably also of use.

 

[JohnMc]

S. cerevisiae is an extremely convenient organism to do genetics on, not least because other people have been doing the hard work for us for the past fifty years. If you're willing to devote a fair bit of time, and a small but reasonable amount of money to it, you can easily do classical genetics on yeast in your garage. You won't need any equipment beyond a microscope. It's not trendy, and doesn't sound as cool as directly manipulating DNA, but classical genetics is entirely powerful enough to create and map mutations.

One thing to consider is that lab yeast has been selected to be readily manipulated (heterothallic vs. wild-type homothallic and that a lot of brewing strains are aneuploid; probably don't mate and sporulate as well).
So maybe start with the parent of S288C and look for brewing qualities from there?
I can't remember the designation of the parent strain, I saw it in a review paper once a long, long time ago, but I recall it was a cross of a Carlsberg lager strain with Fleischmann's bread yeast. That should have some nifty genetics to start to play with.