Bootcamp Planning
From MPPWiki
Back to Events Calendar
We plan to organize a bootcamp summer school in August or September 2011, directly preceding or after DNA 17 (which is scheduled for September 19-23 at Caltech).
The planning is at this point very preliminary and tentative. Notes and thoughts below. MPP faculty should feel free to edit all this stuff below -- for now, everything should be interpreted as having a question mark. Note that Paul started his own subpage with ideas, with a link at the bottom of this page. This is because if two people edit a wiki page simultaneously, one version might get overwritten (although it will still be recoverable from the archive). Feel free to start your own page or edit this one directly.
Contents |
Action items
- Niles: check with Rob Phillips about his bootcamp dates.
- Erik: if (based on Niles' info) there really is a conflict with Rob's bootcamp, see if the DNA computing conference can be moved. (unlikely)
- All: suggest teaching modules in the section below, flesh them out.
Time and location
We're thinking one super-intense week, with lectures in the mornings and labs in the afternoons.
Might want to use the Bi 1X teaching lab in Braun. It has 30 ~8-12 stations, which should be enough if we have 20 participants.
It's set up primarily for microscopy... we'd have to move in AFM, or use other room for other instruments, like spectrofluorimeter, etc.
We expect Rob Phillips will be using it for his bootcamp (Sept 6-17th) two weeks before term starts (Sept 26).
So, to use that lab, we'd need our week to be Aug 29th - Sept 2nd. Can DNA17 be moved?
- RMM, 6 Aug: The lab is not quite as big as the original numbers. There are 4 rows of benches arranged in 2 U's, plus a third row that is dedicated to gels and instruments. You can 16 students in comfortably (this is how big Bi 1X is) and possibly 20 if you use the bottoms of the U's.
- RMM, 6 Aug: The microscopes will not be easy to move and there is not really room for any more equipment in that space, unless we take over bench space
Goals for information transfer
We'd like to focus on principles and enabling people to do things.
- Who's our audience?
- Should we focus on PhD-level participants -- senior grad students, postdocs, faculty?
- Should we aim for folks with backgrounds in engineering -- CS, EE, robotics, controls -- for whom the mathematical foundations won't be challenging, and the labwork will be new?
- Should we aim for folks who are going to do something with what we tell them, e.g. chemists and biologists and biophysicists who could use DNA nanotech & molecular programming to great effect in their research, but the activation energy (without the bootcamp) is too high?
- Understanding DNA
- Chemistry & biochemistry primer for nucleic acids.
- Molecular modeling.
- Thermodynamic and kinetic models (and software) at sequence & secondary structure level.
- System-level programming, sequence design & analysis, compilers
- Chemical kinetics math and modeling. ODEs, stochastics.
- Fundamentals of computer science, models of computation.
- Abstract models for molecular programming -- tiles, CRNs, reaction graphs
- NUPACK analysis & design, Pepper & Multistrand, Xgrow.
- Using instruments & lab techniques
- AFM, fluorescence microscopy, fluorescence spectroscopy, UV spectroscopy, gels, cloning.
- Experience with programmable systems
- HCR, hairpin assembly/disassembly, strand displacement circuits, tile assembly, in vitro enzyme circuits, origami, in vivo something, motors & walkers.
- Presumably, one system each day! That means we'll need 5 topics, 5 experiments... or they could be in parallel, overlapped...
Teaching module ideas
The idea is to teach the above concepts, not thoroughly, but through a hands-on experience.
We might take a project oriented approach. Each project module comes with associated principles lecture and dry exercises (e.g. software or modeling). After a preplanned experiment, an interesting option would be to have a brainstorming session where new experiments (feasible given the constraints) are proposed and then (maybe a few days later, if unpurified DNA orders are required) performed.
Although the general flow of the week will be lectures and dry stuff in the morning, experiments in the afternoon, we will probably need to flexibly allow for variation since some projects will by multi-day and will require some morning experiments. It will be tricky to schedule if there are parallel projects going on simultaneously (participants having in advance signed up for their own selection from our menu), but that may be worth the effort. If we follow Rob Phillip's format, each day will conclude with short presentations by each team, showing what they found out that day.
It would be great to have maybe 10 modules here to discuss.
- Klavins: stochastic chemical kinetics (dry)
- Egbert: tuning genetic regulatory circuits in vivo (wet)
- Soloveichik: strand displacement circuits (dry and wet)
- Winfree: tile self-assembly (dry and wet)
- Goal: grow patterned crystals and tubes from a seed
- Principles: algorithmic self-assembly, analysis of growth and nucleation errors [2 hours of lecture]
- Prepared in advance: an interesting tile set, purified DNA tiles, DNA origami seed, set of tile adapter strands
- Dry exercises: xgrow simulations for a various subsets of the tile set [2 hours of work]
- Wet experiments: each team selects a subset of tiles & seed, anneals w/ and w/o seed, does AFM, analyses data. [6 hours work, uninterrupted]
- Extension: design a new DNA tile to modify the growth process, order it, try it out [2 hours design; 1 day wait for synthesis; 1 day purification by unfortunate Caltech grad student; 6 hours to try it out]
- You: exciting topic here!
Teachers
It will probably make sense to design the curriculum around who we get as enthusiastic teachers.
We might want to follow Rob Phillips's model, and only have a few invited colleagues to help teach, and to choose projects that are of current interest to them, so that the bootcamp ends up giving stimulating their regular research by focussing a bunch of fresh minds on it for a week? (And since there are six of us, we might not invite any outside colleagues...)
- MPP PIs and especially grad students & postdocs.
- Outside experts, if we can get them, e.g. William Shih, Shawn Douglas, Luca Cardelli, Bernie Yurke, Georg Seelig, Vincent Noireaux, Yannick Rondelez, Gerry Joyce, David Baker (and/or any of their students)
