SynBERC:COG/October 2007 Meeting

=Notes=

What do people want to get out of the meeting?

 * Reshma would like to see us work out benchmarks such as: total volume of orders, how fast a turn-around time is needed? As well as to address specific questions like 'would NEB sell a Biobrick cloning kit?'
 * Dueber is interested in things automated assembly allowing him to do things he can't currently do, rather than just outsourcing things he does already. This might also make the turn-around time less of a big deal.  He would be willing to wait for something like a combinatorial library of parts that would be hard to make by hand.
 * George commented (and Randy agreed) that we might be able to do that pretty easily if you didn't mind your library being a pool (rather than individual library members in separate wells) using current approaches.
 * George mentioned that synth companies do cloning at many stages in the process.
 * Randy brought up that the Registry might offer other services besides assembly, such as Biobricking new parts.
 * Dueber pointed out that making parts is really easy, putting them together is the real challenge. He said that Berkeley team was about to make 70 new parts in the first few weeks of the summer.
 * Leonard said that the company he worked at did a lot of cloning (including combinatorial) and he would be willing to bring someone in to talk about their methods with us.
 * TK suggested that we also look at the idea of a Biobrick cloning kit and manual containing important components.

Debriefs

 * TK debriefed on the QIAcube. It sounds like overall it's unlikely to work well for the distributed assembly approach as it has limitations on heating temp, pipetting volumes, among other things.

Homework
Proposed homework for each of the groups for next week: Randy also proposed some homework for the meeting after next (when the Registry will be presenting).
 * 1) What's the cloning amount (and type) per lab ?
 * 2) What experiment can't we do that we'd like to do?

=Agenda=

What are the goals for COG?
Proposed based on SynBERC meeting / conversations since:
 * 1) Stay up to date with current automated cloning research projects taking place in SynBERC labs
 * 2) Get the ball rolling for solving the automated cloning problem in the short term

Research Updates

 * 1/2 the meeting?
 * Brief updates from each of the groups? one longer presentation each month about one project?
 * Something else?

Getting the ball rolling for solving automated cloning

 * 1) Define the problem we want solved
 * 2) Specify the possible solutions
 * 3) Evaluate / Act on solutions

Define problem(s)

 * Combinatorial libraries of assembled parts (Wendall, UCSF iGEM team project)
 * Automation of standard BioBrick Assembly (Registry)
 * Automation of assembly independent of BBs (Synth companies?)

Specify Possible Solutions

 * 1) Decentralized approach
 * 2) *We make cloning easier by using some automation combined with "normal" individual cloning approaches. For example, each lab buys a Qiacube to partially automate minipreps and restriction digests.  Then ligations and transformations are done by hand at the bench.
 * 3) Centralized academic facility
 * 4) *We find the resources to start an assembly facility similar to how sequencing facilities are set up on campus
 * 5) Outsource to DNA synthesis company
 * 6) *We work with a DNA synthesis company so that they can offer commercial automated cloning services. How do we ensure that such a partnership wouldn't fall apart as soon as a big synthesis order comes along?  Automated cloning may be too small a market and/or too large a problem for DNA synthesis companies to tackle right now?

Evaluate / Act on Solutions

 * Registry hires more staff to do process engineering
 * Trial QIAcube / other "cheap" robots for distributed approach
 * Figure out how big an assembly order iGEM + labs using parts could put together for a Synth company (George suggestion)
 * Go talk to Synth companies see if they'll do this.
 * Plot the cost of synth vs. assembly over time for particular length scales (drew's suggestion)

Meeting Frequency

 * Monthly?
 * Any weekday's particularly bad for certain groups? (Weds at MIT)