Topics Topics Help/Instructions Help Edit Profile Profile Member List Register  
Search Last 1 | 3 | 7 Days Search Search Tree View Tree View  

Visit The Brewery's sponsor!
Brews & Views Bulletin Board Service * Brews and Views Archive 2006 * Archive through February 15, 2006 * Evolution of the genome, selection of regulatory strategies - question for Chris Colby or the other bio experts? < Previous Next >

  Thread Last Poster Posts Pages Last Post
  ClosedClosed: New threads not accepted on this page        

Author Message
 

Fredrik
Senior Member
Username: Fredrik

Post Number: 2926
Registered: 03-2003
Posted From: 62.20.8.114
Posted on Friday, February 10, 2006 - 01:12 pm:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

Disclaimer: This is maybe < 5% beer related, and 95% irrelevance.

This is a pretty foggy question but in the progress of working the simulation model, I have faced the regulation problem, which is of course the major one. This make me think.

Given the need for regulation in an enzymatic process, one can consider two modes of regulations, either synthesis and degradation of the enzyme, or various kinds of allosteric or competitive regulations.

It seems to me that for vary dynamical processes, and in particular where the enzyme complexes are big and expensive, a regulatory strategy based only on synthesis and degradation sounds like a very expensive one.

If possible, allosteric regulators by means of simpler, less expensive molecules should in some senses make more sense and be more economic?

Now, I am curious if there is any evidence in history of biology at any level where selective pressure has replaced expensive regulatory strategies for more clever ones, as they are "found", supposedly often by accident?

Also, is there any evidence that the number of effective genes are reduced, as an organism evolves? It seems to me that if any organisms can evolve, and do that same "job", and encode all that in a smaller set of genes, it should be an approvement? Or "simplification"?

Any historic input on this would be interesting.

/Fredrik
 

Fredrik
Senior Member
Username: Fredrik

Post Number: 2927
Registered: 03-2003
Posted From: 62.20.8.114
Posted on Friday, February 10, 2006 - 01:17 pm:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

Though, when I think of it, there is probably again a balance. Because taken to exremes, perhaps a very high compression encoding would be more sensitive to data corruption(mutation)? So not to take it too far, perhaps a certain amount of redundancy is necessary too to resist the noise.

/Fredrik
 

John Thompson
Intermediate Member
Username: Jt100

Post Number: 429
Registered: 04-2002
Posted From: 216.136.22.138
Posted on Friday, February 10, 2006 - 05:45 pm:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

Do the chickens have large talons?
 

ELK
Senior Member
Username: Elkski

Post Number: 1436
Registered: 01-2003
Posted From: 67.177.25.240
Posted on Friday, February 10, 2006 - 05:47 pm:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

Fredrik, the politicos here in SLC are trying to force the schools to say that evolution is just a unproven theory just like intelligent design.
http://www.sltrib.com/utah/ci_3490184

http://mm.isu.edu/pipermail/evolidaho/2005-March/000181.html
 

Marc Rehfuss
Junior Member
Username: Marc_rehfuss

Post Number: 29
Registered: 03-2003
Posted From: 169.237.122.223
Posted on Friday, February 10, 2006 - 06:49 pm:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

This is not directly related to Saccharomyces, but I should mention that bacteria, especially ones with small genome sizes, regulate protein synthesis and transcription without the need for regulatory proteins. This is the process of phase variation. In a nutshell, dinucleotide repeats (ie. CT repeats) or poly(N) tracts (e.g. polyA or polyG) can cause slipped strand mispairing of DNA polymerase. If these polyN tracts or CT repeats occur within the promoter region between the -10 and -35 sites, they can affect RNA polymerase binding, thus affecting transcription levels of the downstream gene. The length of the repeat or poly(N) tract can either increase or decrease transcript levels. (This is transcriptional phase variation) If the poly(N) tract or CT repeat occurs within the ORF (opening reading frame) and results in a frame shift, a premature stop codon can be introduced, resulting in a non functional, truncated protein.

Essentially, this method of regulation is DNA sequence based and highly effective for organisms with small genomes and thus less regulatory proteins. I'm not sure if this occurs in Saccharomyces, since its genome is relatively large (~13,000 KB) compared to bacteria, especially the one I study, H pylori (1,600 KB). Yeast can afford complex regulatory strategies, but I suppose it's possible this occurs in yeast as well.

So yes, selective pressure has resulted in more efficient transcript/ translation regulation in organisms with small genomes.
 

J. Steinhauer
Advanced Member
Username: Jstein6870

Post Number: 838
Registered: 03-2002
Posted From: 216.70.45.1
Posted on Friday, February 10, 2006 - 08:19 pm:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

Fredrik, I'm no biologist, but environmental pressures in very many instances lead to more complex genomes, even in simple organisms. Exposure to vancomycin leads to incorporation of genes (VanA, etc.) in Enterococcus sp. to give them resistence. Similarly, this occurs in Staph aureus resistent to oxacillin (methacillin). Anyway, for more complex organisms to evolve from the more simple organisms, this would have to be the case most of the time. Otherwise, we would be left with only simple organisms. I think the more efficient (simpler) you get, in an open system, the more vulnerable (less adaptable) you are to insult. Having two sets of genes is better than having one, of course.
 

Chumley
Senior Member
Username: Chumley

Post Number: 3914
Registered: 02-2003
Posted From: 65.102.123.53
Posted on Friday, February 10, 2006 - 09:54 pm:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

>>Disclaimer: This is maybe < 5% beer related, and 95% irrelevance.

I would say .005% beer related, 99.995% irrelevance.
 

Fredrik
Senior Member
Username: Fredrik

Post Number: 2928
Registered: 03-2003
Posted From: 213.114.44.246
Posted on Friday, February 10, 2006 - 10:44 pm:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

Thanks Marc and Steinhauer! Interesting. I guess there is a logic to the different variations in each case as "efficient regulation" could be measured in different ways. At first I was referring to "efficient" is in minimizing overhead so to speak, but like Steinhauer notes, reducing overhead at the expense of flexibility and increase vulnerability is not "efficient" either.

I think I need to dig deeper into this to understand it better. I'll consult my textbook. What I am after is to get a grip on where the overhead possibly is, in terms of "time/latency" to responses, but also in terms of cost as in energy and building blocks. ie. the cost of the tools.

The idea I have in the model is to first of all simplify because otherwise it's way too complex, but to try and reduce the regulatory model down to a level where the energy and time overhead is somewhat neglectable relative to the time and energy scale of the overall metabolism. Then at this level, I will try to implement an ad hoc optimizing routine to mimic cell intelligence. But that would be invalid if the energetics and response times at that level is significant.

Thanks alot for the comments!!

/Fredrik
 

Ken Anderson
Senior Member
Username: Ken75

Post Number: 1380
Registered: 11-2002
Posted From: 69.168.141.10
Posted on Saturday, February 11, 2006 - 01:26 am:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

What Marc said.
 

ELK
Senior Member
Username: Elkski

Post Number: 1441
Registered: 01-2003
Posted From: 67.177.25.240
Posted on Saturday, February 11, 2006 - 01:57 am:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

Marc, do you know much info on transmission methods and health effects of H. pylori?? Can it be transmitted in homebrew of some aspect of tasting SG reading or eating raw grain??
That is wicked stuff.
 

Sean Richens
Intermediate Member
Username: Sean

Post Number: 280
Registered: 04-2001
Posted From: 142.161.37.34
Posted on Saturday, February 11, 2006 - 04:21 am:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

As organisms get more complex overall, more of the phenotypic variation is through regulatory mechanisms of one sort or another.

One effect of this is that organisms get to carry around inactive genes as a bank of randomness which affords some protection over the generations against disease or even natural disaster.
 

Matt Bobiak
New Member
Username: Aeneas

Post Number: 8
Registered: 11-2005
Posted From: 69.123.65.241
Posted on Saturday, February 11, 2006 - 05:10 am:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

Without getting into genome stucture, as that's not my field of expertise, genomes rarely decrese in complexity as an organism evolves. Many have heard that much of the human genome was considered "junk DNA" with no coding or regulatory functions. Within the past several years, new types of gene regulation and regulation of protein synthesis have been attributed to this "junk DNA' and have likely served critical functions in our evolution as mammals, chordata and possibly multicellular eukayotes.
I am unfamiliar with your modeling project you discuss but metabolic modeling is often based on energy potential within a cell. These models are based on the abundance and turn-over rates of ATP, GTP, NADH+ and FADH2. The abundance of these factors is what really gets to the bottom of enzyme regulation, because most enzymatic reactions require at least one of these. If you plan on modeling the general enzymatic level, I would suggest that energy potential would be an acceptable parameter, if however you want to model EVERY reaction, and how the kinetics, protein abundance and substrate abundance of each reaction change in regards to an stimuli and how they change in regards to each other, we should be speaking on another, non-homebrewing based discussion board.
Matt
 

Randy McCord
Advanced Member
Username: Mccord

Post Number: 574
Registered: 02-2003
Posted From: 216.174.177.202
Posted on Saturday, February 11, 2006 - 07:00 am:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

HUH? Where am I? What?

Oh yeah, BEER, I think I need another. Or maybe I've had enough? I'm confused.

(Message edited by mccord on February 11, 2006)
 

Fredrik
Senior Member
Username: Fredrik

Post Number: 2929
Registered: 03-2003
Posted From: 213.114.44.246
Posted on Saturday, February 11, 2006 - 10:13 am:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

Yes Matt, the model is definitely getting comlpex, and while the desire for the model, is partly based in curiosity of a homebrewer, the technical details is probably not overly relevant to this board.

Yes, I will definitely have energy and redox carriers part of the model. Energy and redox balance are clearly two of the main things.

I am somewhat inspired by reasearch on metabolic networks, but some of these models to the optimatio directly on the reaction rate level. I find that to bee too crude, I want to put it one step below, at the enzyme regulation level.

I am going to have a simplified and reduced set of dynamical variables. This is the most important metabolic intermediacte (but reduced, so not ALL of course, but most important). I will next also consider the enzymes to be dynamical variables, and I will attempt modelling their synthesis/degradation. The enzyme level regulations will be put in by simplified standard kinetic math for various allosteric effectors and also there will be possibilit for enzyme competitions. The basic idea is that the effector induces transformations in the enzyme following a reversible reaction, kind of like T and R states.

Next there will be a set of pathways. Each pathway is registred as as type. Every enzymatic pathway have the possibility of sharing enzymes with other pathways, so there I will automatically account for enzyme competition.

I am trying to find a simplified and uniform mathematical expression for an arbitrary rate regulation.

I have classified some of the pathways as "regulatory", the philosophical idea of these, is that the are ultimately the strings of the puppet that the chief of hte cell is pulling *at will* in order to navigate the cell through life. And the chief of the cell has in mind to optimize survival and growth, to the best of his knowledge.

These regulatory pathways are finally controlled by a set of real numbers, that represent signaling molecules that are in charge of overall metabolic expression, ranging from enzyme control to transcription. I will not go deeper, but it's hairy enough as is.

Once I've arrived at a structured and streamlineds strucure, all I need to do is to implement this numerically, and parametrize some of the initial values, find initla values of all variales, and define the goal of the cell chief as a mathematical expression.

Then I will cross my fingers that my computer has the power to execute it withing reasonable time, which I doubt.

Then finally it would be some tuning work.

/Fredrik
 

Fredrik
Senior Member
Username: Fredrik

Post Number: 2930
Registered: 03-2003
Posted From: 213.114.44.246
Posted on Saturday, February 11, 2006 - 10:28 am:   Edit Post Delete Post View Post/Check IP    Ban Poster IP (Moderator/Admin only)

I am actually two things in parallell.

1) Trying to find how to systematically model an arbitrary metabolic network in the best way. In theory this would apply to any organism, not just yeast. This will supply the main mathematical strucutre of the model.

2) The other thing I'm trying to understand is specifically s.cerevisae in a beer fermentation. I'm doing some reading and combining all that into a consistent model. Some of the main processes are the lipid metabolism, the aminoacid/protein metabolis, and the carbohydrates, then also some DNA/RNA stuff. I'm trying to understand the main overall routes, but I also try to see that I don't miss any details that's potentially relevant to flavour formation. Ultimately I want to use the model to elaborate flavour formation.

Like in ester synthesis, I will have to register each known gene and describe the pathway, and it's affinity for the relevant substrates.

Hopefully when all those things are implemented the model will help explain how ester synthesis is really regulated from the brewers perspective.

If it doesn't then it means the model is wrong, and I have thought of that too, because the model is oing to be designed in a hopefully flexible way. So adding or removing pahway could be done from a databse interfece. Just add a new pathway, describe it's stochiometry and basic kinetic type + some guesstimated kinetic data. If the guesstimated kinetic data, like affinity constants, are not know, you have the option ot change the kinetic type to a simpler one, that requires less parametrization and leave it's full regulation up to the cheif of cell! only describing the stochiometry.

/Fredrik