Math, Microbiology, Game Theory, and Beyond

32 views
Skip to first unread message

Giovanni Lostumbo

unread,
Jan 19, 2011, 6:48:09 AM1/19/11
to DIYbio
Hi, I've anyone has an interest in virology and microbial communities
along with mathematical expertise, if they would be able to explain
the/any differences between one or both of the first two papers (and
themselves, though related) and the third one listed below.

http://www.pnas.org/content/107/30/13197.extract
"Bacteria determine fate by playing dice with controlled
odds" (subscription/paywall)

http://www.pnas.org/content/106/50/21027.long
"Deciding fate in adverse times: Sporulation and competence in
Bacillus subtilis"

“Prisoner’s dilemma in an RNA virus.,” Nature,
http://www.nature.com/nature/journal/v398/n6726/full/398441a0.html
(subscription/paywall)

Maybe they can't be "compared" since they're microbes and viruses (and
the former has different species). Nonetheless, if there are any
papers that can analyze the dynamics whether in mathematical terms or
other, such as if it can explain alternate scenarios for game theory
and beyond, I would be interested. Thanks.

J. S. John

unread,
Jan 20, 2011, 5:11:10 PM1/20/11
to diy...@googlegroups.com
On Wed, Jan 19, 2011 at 6:48 AM, Giovanni Lostumbo
<giovanni...@gmail.com> wrote:
> Hi, I've anyone has an interest in virology and microbial communities
> along with mathematical expertise, if they would be able to explain
> the/any differences between one or both of the first two papers (and
> themselves, though related) and the third one listed below.
>

I tried to read them. I'm working my way thru the 2nd one. From what I
read in the 1st one, it talks of signal inputs analogous to binary
switches found in EE. Now, I don't know what game theory is about but
so far I read about stochastic nature of the signals but I don't know
much about that area either. I understand that the signals affect gene
expression.

The 2nd one seems to be an in-depth article whereas the 1st was more
of a summary. I haven't made it to the 3rd yet.

Has anyone else read the papers? What do you think?

moteyalpha

unread,
Jan 20, 2011, 5:31:39 PM1/20/11
to diy...@googlegroups.com
I skimmed them and I understand game theory and the prisoner's dilemma
as well as the Markov chain Monte Carlo methods.
An interesting free opencourseware on game theory is available at Yale here:
http://oyc.yale.edu/economics/game-theory/
The Yale course is interesting , funny and informative on many levels.
If I had to guess about the articles, which is all I am doing, the third
article discusses phage which have a smaller genome and thus less chance
of developing complex response and so they choose poorly because they
have shallow genetics and thus cannot maintain a complex look ahead
response.
It does seem that the first two describe virtually the same effect.

Cathal Garvey

unread,
Jan 20, 2011, 5:35:08 PM1/20/11
to diy...@googlegroups.com
I know a little about subtilis, at least! The strategy in competence is all last-ditch game theory timing by the look of things. The stationary-phase pathways leading to competence lead ultimately to antibiotic production and sporulation, each stage of which carries its own potential advantages.

For the B.subtilis colony considering its options, when times are tough a smash-and-grab from the locals is a viable way to improve matters sometimes. Perhaps the food supply is being eaten by a competing bacteria, so antibiotics offer a competitive edge and (also murder a few siblings for food and other cells for DNA that might enhance metabolic prowess). If a competent cell is lucky enough to find some scraps of DNA that improve matters, it doesn't need to sporulate after all for a while, and it can swarm on to a brighter future! If nothing else, antibiotic-lacing the soil is a nice way to improve odds of the spores being left alone.

In-vitro, the timing of events would be expected to get deregulated, because the order of events is more predictable and the selective advantage of having a "game theory" approach to your surroundings is less important. Indeed, competence development often suffers in-vitro, with species losing competence over time. I believe this was found true of A.baylei ADP1, and I believe it's true of B.subtilis likewise (although 168, probably one of the most domesticated strains of bacteria ever, is still competent.. so perhaps it's not a universal rule).


--
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To post to this group, send email to diy...@googlegroups.com.
To unsubscribe from this group, send email to diybio+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/diybio?hl=en.




--
letters.cunningprojects.com
twitter.com/onetruecathal
http://www.indiebiotech.com

Giovanni Lostumbo

unread,
Jan 20, 2011, 11:21:22 PM1/20/11
to diy...@googlegroups.com
"If I had to guess about the articles, which is all I am doing, the
third article discusses phage which have a smaller genome and thus
less chance of developing complex response and so they choose poorly
because they have shallow genetics and thus cannot maintain a complex
look ahead response."

Without reading the paper (I don't have access to the third article),
and with my bio background, it too was my suspicion that viruses'
actions in this study are "less complex" than microbes, perhaps
because of their evolutionary phylogeny and "dimer"(?) molecular
dynamics, and their inability to signal (or signal diversely), such as
with quorum sensing in microbes, which affect gene and thus protein
expression (real time), with more variables and members involved in
the "game."

"It does seem that the first two describe virtually the same effect."

That I might disagree with as I suggest and with further:
For the first two papers, it was covered in the news here:
http://www.physorg.com/news179521562.html
"But how they make this decision and which cells take this chance have
been a mystery."
[above quote: my guess here is it's explained by quorum sensing/communication]
"The researchers discovered in their study that the bacteria’s game
theory decision making process is far more advanced than the
well-known game theory problem known as the Prisoner's Dilemma."
..."Because the number of participants in a bacterial colony can be up
to 100 times the number of people on earth, the bacteria need to
construct a more complex form of game theory. The rapidly changing
environmental conditions they face means also bacteria have limited
time to decide."
It also suggests a positive interpretation from bacteria:
"According to Onuchic, bacteria usually do not cheat their friends and
inform them by sending chemical messages about their true intensions."
http://www.physorg.com/news/2010-10-microorganisms-lessons-gamblers-rest.html
http://www.sciencedaily.com/releases/2010/10/101012121439.htm

These articles, if they've been accurately reported from
correspondence with the authors of the papers, seem to suggest
"social" applications, which might not translate favorably, but are
probably more interesting/useful/sophisticated than existing game
theories being applied/tested in social science, economics, and other
fields today. I'm not sure if there are many other papers analyzing
these dynamics, but it seems like there are multiple ways to interpret
the decision making processes of diverse microbial communities, taking
into account quorum sensing, chemotaxis, and horizontal gene transfer.
Thanks for all your responses.

moteyalpha

unread,
Jan 20, 2011, 11:35:36 PM1/20/11
to diy...@googlegroups.com
That is very insightful and interesting.
The idea that a far more complex social dynamic is at work than exists
in the culture of 6 billion more complex agents.
The quorum sensing aspect is one of my favorite studies and has meaning
across many different biological systems.
The self assembling structure of fungi is the one which intrigues me
most as the fact that they live as essentially free agents and on a
trigger like the depletion of nutrients form themselves into a somewhat
complex multicellular structure that propels itself to a new environment.
It is interesting to contemplate that an intelligence or collective can
exist in such a distributed form.
The structure of bot nets is a place where that same strange situation
exists such that an agent has purpose and effect and yet has no real
physicality in the normal sense.

On 01/20/2011 10:21 PM, Giovanni Lostumbo wrote:
> "If I had to guess about the articles, which is all I am doing, the
> third article discusses phage which have a smaller genome and thus
> less chance of developing complex response and so they choose poorly
> because they have shallow genetics and thus cannot maintain a complex
> look ahead response."
>
> Without reading the paper (I don't have access to the third article),
> and with my bio background, it too was my suspicion that viruses'
> actions in this study are "less complex" than microbes, perhaps
> because of their evolutionary phylogeny and "dimer"(?) molecular
> dynamics, and their inability to signal (or signal diversely), such as
> with quorum sensing in microbes, which affect gene and thus protein
> expression (real time), with more variables and members involved in
> the "game."
>
> "It does seem that the first two describe virtually the same effect."
> That I might disagree with as I suggest and with further:
> For the first two papers, it was covered in the news here:
> http://www.physorg.com/news179521562.html
> "But how they make this decision and which cells take this chance have
> been a mystery."
> [above quote: my guess here is it's explained by quorum sensing/communication]

> "The researchers discovered in their study that the bacteria�s game

Inigo Howlett

unread,
Jan 28, 2011, 10:27:15 PM1/28/11
to DIYbio
You sound like just the person to write the article.

On Jan 19, 6:48 am, Giovanni Lostumbo <giovanni.lostu...@gmail.com>
wrote:
> Hi, I've anyone has an interest in virology and microbial communities
> along with mathematical expertise, if they would be able to explain
> the/any differences between one or both of the first two papers (and
> themselves, though related) and the third one listed below.
>
> http://www.pnas.org/content/107/30/13197.extract
> "Bacteria determine fate by playing dice with controlled
> odds" (subscription/paywall)
>
> http://www.pnas.org/content/106/50/21027.long
> "Deciding fate in adverse times: Sporulation and competence in
> Bacillus subtilis"
>
> “Prisoner’s dilemma in an RNA virus.,” Nature,http://www.nature.com/nature/journal/v398/n6726/full/398441a0.html

Giovanni Lostumbo

unread,
Jan 28, 2011, 10:42:50 PM1/28/11
to diy...@googlegroups.com
:) Thanks Inigo. I would gladly like to write an in-depth analysis of the papers and current advances in this particular field if I had the time. There are millions of bacterial species and its possible many different social network patterns (or not) exist and are likely to offer options to interpretations of game theory, and so understanding the multiple types of network behaviors/actions is as important as the theorems that might be singled out simply because it's been already established. So more diverse microbial species' observations of community dynamics here still has lots of alternative explanations to offer to existing game theories, in addition to building new ones from scratch.

Reply all
Reply to author
Forward
0 new messages