ETH 2006 Ideas

From 2006.igem.org

Revision as of 14:35, 27 July 2006

back to ETH 2006

Last years brainstorming: Previous_Ideas

Cellular automata / Quorum controlled behaviour

• Game of Life or other semitotalistic automata
  • [http://en.wikipedia.org/wiki/Conway%27s_Game_of_Life wikipedia]
  • [http://csb.inf.ethz.ch/igem-2006 life demo]
  • [http://www.ics.uci.edu/~eppstein/ca/lifelike.html others]
• propagate a pulse in culture
• Predator-prey-veggies (or just predator-prey)

State representation:

• switches and reporters
• cell life (self destruct and regenerate)

Stepping could be done by

• an exogenious signal, so each step is a process converging to a steady state (avoid problem of developing a dynamic system)
• generations
• oscillating system in cells

(Normally cellular automata are stepped synchronously, but it's often not really necessary)

The matrix would not be as regular as in the mathematical models of corse ...

Logic

→ see [http://en.wikipedia.org/wiki/Logical_gate logical gates in wikipedia]

General Notes

Can't find any logic gates except inverters in the registry, looks interesting.

→ because of that, even simplier gates (AND, OR, XOR, ...) are required

→ every gate on the way to the project goal is

a) of great value for the registry

b) an intermediary step (and success) for the project

Simple ALU

A kind of a simple ALU: at least two simple logic functions and a multiplexer

• [http://en.wikipedia.org/wiki/Arithmetic_logic_unit alu in wikipedia]

Half-adder or Full-adder

An implementation of a half-adder or full-adder: 1-bit adder with carry

• [http://en.wikipedia.org/wiki/Full_adder half/full adder in wikipedia]

Idea: Pattern recognition

Main Goal:

• Write something with a chemical on a petri plate (like ETH for example)
• Let Bacteria grow uniformly on the plate
• Expose the plate to a picture (black and white) of the same pattern
• Result:
  • Bacteria gets green when pattern on the plate and picture match (light and chemical)
  • Bacteria does not express fluorescent protein when pattern on the plate and picture match (no light and no chemical)
  • Bacteria gets red when pattern on the plate and picture do not match

Implementation from the ingenieur point of view

• Half Adder
• = opportunity to build an AND and an XOR Gate

Implementation from the biologist point of view

• AND Gate: presence of chemical induces the production of an inactive protein, which is activated by a second protein (produced when light is present): the first protein is then activating GFP production
• XOR Gate: I think Marko had a good idea about that (I didn't really got it)
• a light sensible promoter (there are some)
• a chemical sensible promoter (chemical to be defined)
• we would need to give a kind of first signal to the system (to don't have light reactions before a certain moment, otherwise we have to work in a dark room...), an activation signal after the bacterias have grown on the plate

Illustration of the above concepts: media:Half-adder.pdf

An irreversible switch

An irreversible switch (lambda phage like)

A/D converter

analog digital converter, somewhat simplified, e.g. like this:

• add some chemical be recognized by the cells, change the concentration over time (with some defined frequency)
  → this is our analogous signal to encode
  note: If a PoPs input is used, the A/D converter is more valuable since independent of the measurement device --Ajk 15:46, 23 July 2006 (EDT)

• cell reaction, depending on chemical concentration:
  1. too low: no reaction
  2. low: reaction α, e.g. green fluorescent
  3. medium: reactions α and β, e.g. red and green fluorescent
  4. high: reaction β only: red fluorescent
• implementation notes
  • the green fluorescent protein is the reporter of a band pass filter, i.e. low and high concentrations don't pass the filter (no green)
  • the red fluorescent protein is reporter of a high pass filter, i.e. high concentrations cause the production of the protein
  • the lower boundary of the red filter is above that of the green filter
  • to enable encoding of a signal, that is of a sequence of different concentrations, the fluorescent proteins have to be degraded quickly. this might be possible by attaching degredation recognition patterns to the proteins, but is yet another challenge. however, this could be the last "bonus" step of the project.

Sensing

What?

• acoustic
• pressure
• electric
• magnetic (exists in nature, cf. [http://magnum.mpi-bremen.de/magneto/pub/Schueler1999.pdf] and [http://www.chemcases.com/cisplat/cisplat01.htm])
• light (BB exists[http://partsregistry.org/Featured_Parts:Light_Sensor])
• temperature (first BBs by 2005 Toronto team [http://parts2.mit.edu/r/parts/partsdb/pgroup.cgi?pgroup=iGEM&group=iGEM_Toronto])
• chemicals, metals
• number of neighbor cells (related to «Game of Life» idea): it seems that another team tried similar in 2004 [http://2006.igem.org/wiki/index.php/IAP2004:Polkadorks]

Agents (sense & act)

• fire squad: get bacteria to find a reaction on the plate and extinguish it
• bacteria sensing high atmoshpere and switching to produce ozon (to refill ozon holes)
• extract sun milk when needed (depending on sun intensity & skin properties :-) )

Multi-level sensing

• a system which has more than on/off levels, e.g.
  • does nothing if ligand is not present
  • turns green on low concentrations
  • starts stinking (e.g. by producing ammonium) on high concentrations

System that potentially could sense any chemical substance

• leucine zippers that could only build dimer with a bridge
• chemical substance that should be detected is bridge
• the bridge dimerization could be realised by fusionate an original part of a leucine zipper monomer with the antibody that is generated for the chemical substance
  • If we can get two different recognition side for antibody and antigen this could work
  • mice could generate the searched antibody, but for test we could use already described antibodies.

Filamentous Growth

• Make use of the Yeast Filamentous Growth to create patterns with E.Coli (a flower, a line, a fractal)
  • Under certain conditions (lack of nutrients and mating pheromon present) certain yeasts shift to filamentous growth where the cells become more elongated
  • Yeast colonies form then "tree like structures" that are quite beautiful
  • We could implement this in E.Coli (E.Coli can switch to filamentous growth in presence of magnetic fields [http://magnum.mpi-bremen.de/magneto/pub/Schueler1999.pdf] and [http://www.chemcases.com/cisplat/cisplat01.htm]) to have it reacting in this way to other chemical
  • We could use it to create patterns in the Petri plate with E. Coli.

A mini multicellular organism

3 types of cells:

• the sensors
  • sense a chemical in a solution
  • warn other sensors that there is a chemical in the solution (feedback loop to activate sensors)
  • activate cleaners

• the cleaners
  • warn other cleaners that there is chemical in the solution (feedback loop to activate cleaners)
  • clean the solution (get rid of the chemical)

• the regulators
  • regulate the balance between the population of the cleaners and the sensors (so if there is chemical in the solution more cleaners, otherwise by default more sensors)

Interesting from the biological point of view (system is working as a mini multicellular organism) and from the ingenieur point of view (as each of the cell types is made of a half adder (so AND and XOR Gates))

If one of you gets enthousiastic with this idea mail me, as I will not be present on thursday. Then I could developp the idea a bit more (Alexandra).

Oscillation related

• frequency discrimination
• synchronize oscillations among cells (or have emergent synchronous oscillations in the culture)
• frequency divider / multiplier

• use the mechanism developped in Dictyostelium discoideum (D.D.) (an Amoebae: a unicellular organism).
  • when the food supply is exhausted, D.D. aggregates to form a multicellular assembly: a fruting body (to spread farther in the environment)
  • this aggregation is controlled by cyclic cAMP release (cyclic cellular automata)
  • [http://www.springerlink.com/(w2aabf45xs2m2n55rlyffo55)/app/home/contribution.asp?referrer=parent&backto=issue,9,14;journal,2,54;linkingpublicationresults,1:119979,1|Dictyostelium: A Prototype for Spatio-Temporal Organization and Pulsatile Intercellular Communication]
  • [http://www.nature.com/embor/journal/v7/n7/abs/7400714.html;jsessionid=86A9B2B2A6CD6EB4C7D6691BEC50CC68| Transcriptional regulation of Dictyostelium pattern formation]
  • we could for example make E. Coli grow in a certain fashion (a line, a flower or a fractal) using this system

Limitations to lift

• E. coli can't move (Seems Penn state had moving E. Coli last year: Penn_StateProjectDes)
• need to control growth to stabilize cell culture structures if they are desired

Misc

• optical lenses: program different (light) refraction angles
• fractals: let the bacteria culture grow such that fractal patterns appear [http://classes.yale.edu/fractals/Panorama/Biology/Bacteria/Bacteria.html]