IGEM:Harvard/2008/Brainstorming

Click here for full brainstorming board image

We have a lot of good ideas to start working off of! Please add any ideas you have in this space, and make sure to check out the papers here

Theme

We're interested in creating an interface between living systems and electronic systems using bacteria that naturally produce electricity. This idea can go in many different directions depending on what kind of sensors we use, how we process the inputs, and how we present the output.

• Is this a definite theme? It's just that I see a few problems with the biology/electronics interface. For the CCD, for example, will each "pixel" be able to have a linear measure of the brightness? I suspect that noise would really hamper that. If we apply a threshold of some sort, though, the bit count won't be able to be much higher than 2 (or is there some genetic way of applying more complex thresholds?). Also, won't there have to be additional noise processing of the signal output from the bacteria? I like the idea of playing with electricity, but I'm worried that the noise processing will take more time than the biology. -MXH
• I think there could be other themes, but it seemed like this was the main thing that came out of the meeting. I think figuring out how to deal with the noise in the system will be an interesting part of this (and probably any) project we choose.
  • Forgive me for being ungrateful of the advisors, but I feel that a lot of our ideas have revolved around electrical systems simply because at the first meeting we (the students) didn't really have a chance to talk, and the CCD idea was the one the faculty seemed to like. I'm not saying that we shouldn't consider the faculty's ideas (the lava lamp that Prof. Hansel suggested is one I really like), but I think it would be nice if we were forced to think a bit more on our own. It seems to me that we don't talk much when the faculty are around, so they understandably can only try to keep presenting more of their ideas. Maybe the reason is that everyone else wants to do an electrical/biology interface project, in which case my point is invalid. As I've put forth, I would rather play more with biology than with widgets, but where does everyone else stand? -MXH
  • You should speak up more at meetings! It's hard to know what people want when they're not saying anything. Everyone is happy to hear your ideas, especially if it's different than what we are saying :)

Other themes to develop:

• Color
• Plastics
• Drugs

Bioelectricity and Nanowires

• Species
  • Shewanella
    • grows easily, is well characterized, and is similar to E. coli
  • Geobacter
  • Synechococcus
    • Photosynthetic
  • Pseudomonas
• How do we harvest/detect the electricity?
  • building an anode in a plate format

Synthetic Circuits

• Cyborg
  • Bacteria that follow a maze made by other bacteria
    • I like this idea (and no it was not mine)- I think it can incorporate a lot of the other ideas in a modular fashion that still allows us to get an end product even if not all the modules work. -MXH
      • One strain forms "walls", another strain grows between walls
        • For those people interested in chemical "painting"- we could chemical paint the walls and have the wall strain growth exclusively over the chemicals and then have the other strains be repelled by the the chemical. I don't think this is that trivial, since we can't have the walls growing until they close up the path. The could also work for people who want bacteria to detect drugs- we could paint in steroids...
        • To generate a moving blob of bacterial going thru. the maze, we could add some sort of chemoattractant to the entire plate but also make the bacteria rapidly degrade the substance. This way, a given blob will always try to move forward towards the higher source of the chemoattractant and would not be usually turn back towards the regions where it has already degraded all of the chemoattractant. In additional, it might be nice to create another layer of complexity by making one strain prone to following magnetic field lines. We can then place a magnet at the end of the maze and see if this helps this strain or ends up attracting it towards a place where it becomes trapped. E. coli exhibits chemotaxis (see this paper). The same genetic circuit could be used to trigger cell death in the absence of the chemoattractant. Additionally, the chemoattractant could also trigger increased cell growth (or can bacteria not grow any faster?) to make the leading edge of the blob move even faster. Admittedly, this might just make the blob stretch out into a long smear.
        • If we're interested in signals, stochasticity, or colors, I think we could create a really cool system whenever a blob of cells encounters a T-junction. Since the blob would then start growing in both directions and eventually split into two blobs, it would be really cool if we could have cells at the junction walls that trigger the bacteria growing in opposite directions to fluoresce in different colors. For those who're interested in the lava lamp idea, wouldn't it be nice to have the blobs change color? I think there is some hope in this: here's a paper that may be useful.
        • The product should be extremely visually satisfying: a blob of cells that all appear one color grows and repeated splits into different colored blobs at maze branches. Each blob keeps growing, and we can track all the different blobs and see which one wins (including a subset of the initial blob that may be primed at first by magnetic fields). Additionally, this way we have lots of modules that are all nifty, but would not all need to work. At worst, we could just have walls and the bacteria grows thru. all the regions between the walls. It shouldn't be that hard to make the bacteria grow as a moving blob, so we should be able to implement that. Even this, without all the colors and magnetic sensing systems, could be pretty cool.
    • Could we possibly use antibiotics for the wall? Based on the zones of inhibition/resistance of the bacteria to the antibiotics, we might be able to direct the growth causing the bacteria to move forward in the maze (as opposed to also growing sideways). --TA
    • Could we use Bacillus subtilis as the bacterium that goes through the maze? There are many different morphological states (swimmer, spore, biofilm, etc.) and the genetics behind these pathways is well characterized. So maybe if the bacteria start moving into the wrong part of the maze (e.g. a dead end) then it could change morphology (possibly form spores) and stop growing. --TA
      • • Neat! What would the wall bacteria be excreting in order to block the traveling bacteria?
    • I also thought the maze idea was intriguing. -LS
  • Bacteria that play games
    • Tic-tac-toe (Neat paper about DNA tic tac toe)
      • I thought the tic-tac-toe idea was especially interesting. One possible approach we could take is to signal our well choice with a chemical -> those bacteria produce electricity (and maybe change color) -> send a signal to the computer -> have the computer make the bacteria's choice -> inform bacteria in chosen well w/ an electrical signal (which maybe induces a different color change). With this approach we're really more playing tic-tac-toe with the computer than with the bacteria, but it could be a cute way to, as proof of principal, make bacteria which send electrical signals to a computer in response to a chemical and then maybe make bacteria which respond to electrical signals. -LS
      • Or we could take an approach more similar to that in the paper where each human move is accompanied by the addition of a certain stimulus (or combination) which will activate only one well which could be detected with an electrical signal or color change.
        • Or it would be really cool if instead of people adding the chemicals along with their move, the bacteria did. We could create bacteria which in response to our signal, send out a signal to the bacteria in the other squares, activating only the bacteria in the desired area. I think it would be much more interesting if the bacteria are signaling with each other rather than responding to chemicals we add to induce the correct move by the bacteria -LS
        • I like this idea! Can we do this in one kind of bacteria or do you think we would need "sender" and "receiver" cells?
          • I don't know much about processing logic, but how complex is the algorithm for optimal strategy? Could somebody post it here? Thanks -MXH
          • Here's a sort of long-winded description of the strategy. Basically, you will always tie if both players know the right move, and there aren't that many options for each turn since the board is so small. We can play some games next time we meet :)
          • Also, in the paper above they simplified the algorithm by having the molecules/bacteria always go first and always start in the same place

• • • Programming
  • "Light-bright"
    • Pattern formation
      • I like this idea, especially if we could incorporate it with some of the other ones we've had re: lava lamp and bacterial TV. Here's a paper that has bacteria form a bullseye; I think this is what everyone was talking about? They sort of cheat by getting the sender cells in a certain conformation first, but if we could make the sender cells move in a random blob, or be programmed to form colonies of certain shapes (I have no idea if the latter is possible), and then have the receiver cells interact with them, we could make for some really cool patterns. -AL
    • Bacterial "TV"
• Water Quality tester
  • Chemicals
  • Drugs
    • Steroids
      • Would this have to be done in yeast? -MXH
      • Probably, it might have already been done and we can get that yeast or the part as a component.
  • Acids
  • Quantitation?
    • Hmmm? What does this mean? -MXH
    • It means can we be quantitative with the measure of how much of a certain chemical is in the water or can we only see on/off.
  • How would this system compare to what exists already?
• Other inputs
  • Responding to electricity
  • Magnetism
  • Pressure
  • Sound
• Other outputs
  • Sound
  • Color

Other Ideas To Think about

• Bacteria that make a solid substance so you can "build" with bacteria---BioBricks!
• Bacterial "game of life"
• Bacteria that can live on Venus
• Plastic-eating bacteria
• "Autotrophic" E. coli that can survive on one-carbon sources like methane
• Bacteria that produce therapeutic antibodies
• Bacteria that express viral receptors to lure viruses away from human cells
• Bacterial drug delivery systems
• Bacteria that digest lactose for lactose intolerant people
• Logic gates
  • Either/or
    • Is this the XOR/XNOR gate that was mentioned? Say we want signal C if either A or B is true. C could be GFP that requires active TF1 to be expressed. A and B are both extracellular signals that trigger the expression of TF1. However, GFP expression can also be turned off by the repressor REP1. A activates a transcription factor TF2 that activates transcription of REP1. However, make the transcript of REP1 such that the ribosome binding site is block by the mRNA 2^o structure. B activates transcription factor TF3 that transcribes an mRNA gene encoding a short bit of RNA to make the ribosome binding site on the REP1 transcript become unblocked. Is this too complicated? I'm too tired to think clearly, so I just wrote the first thing I thought of, but I feel using an mRNA gene would make the detection that both A and B are present a bit faster than using more proteins? -MXH
• "Homesick bacteria" Spatial memory
• Lava lamp
  • I wrote a bit on color under the maze idea. For this idea specifically, I remember Pam mentioning something a previous iGEM project where bacteria made gas bubbles that would made them float. This, along with some sort of cell adhesion substance (could we just use B. subtilis matrix?), seems like it would make the lava move up. Maybe we could use a strain that is poisoned by contact with oxygen from a water-air surface but not by the amount dissolved in water (are there such bacteria?) so that once a mass of bacteria moves to the top they would start to die, stop producing the gas bubbles and sink again. This might allow for cyclic lava lamp like movements? Is there any other reliable way of generating cycles? -MXH
  • Maybe we could create some kind of gradient instead of relying on oxygen concentrations?
    • Is there an easy method for maintaining a gradient in a fluid column you had in mind? That would certainly be much more efficient that working with dissolve oxygen concentrations.
    • I know it's easy to make a sucrose gradient, could we make bacteria do different things in response to different sugar concentrations?