Constructing the Course
The content and style of the course will be dynamic, meaning
that it will change as we progress through the various stages.
This is very uncommon IRL because
of the constraints of exams, timetables, implied contracts, etc.
VSNS gives us an almost unique opportuniy to do something radically
different, that will be creative, rewarding and very much in the
spirit of the underlying subjects.
Guidelines
We will set limits to the course material in both the breadth of the
coverage and the depth. At present I don't think we have the time
to get involved (to any depth) in subjects like:
- Mechanisms of Protein Folding
- Molecular evolution and the origin of macromolecules
- Force-fields and computation
- "Drug design"
- Ab initio protein structure prediction
- Enzyme mechanisms
- Political, social and economic impact of biotechnology
(This doesn't mean we shan't mention or discuss them - just that we
have no present commitment to provide resources to investigate them).
If there is a demand, we'll set up a listserv so that those who want
to pursue them can explore in an uinstructured manner. If this is
successful, we could consider future
VSNS courses (but don't
get too eager yet!)
There is, of course, no single textbook IRL that covers what we want to
do, in the way that we want. But the excellent
book by Carl Branden and John Tooze maps
well onto much of our content and will be of similar depth. It will
certainly be referred to by course partcipants.
"Simulated Annealing of Course Material "
(If you don't understand this, don't worry! )
In computing properties of proteins an algorithm called simulated
annealing is often used. We probably shan't deal with it on this
course, but the way in which we develop the course material will
look something like it! At the start of SA optimisation process,
the relevant components are assembled and constraints
about the final outcome are applied. (That's the stage where we are
now (Oct)). Then everything is 'heated up' - things fly around at
great speed and it looks nothing like the final outcome.
(That probably takes us to the start of December. If things look
chaotic then, it's a good thing). Then a steady process of 'cooling'
is applied, and large parts attain their final (near-optimal)
state. Smaller bits continue to change (and we expect small changes
throughout the course). The solution is not guaranteed to be the 'best'
but it should be 'near-optimal'. Needless to say, it's a lot of work
and there are countless interactions between the components.
Dynamic nature of the rest of the world
We can guarantee that during the course many things will happen
which change our view on how we see things now. Here are
a few:
- Better interactive WWW/multimedia technology
- New protein databases and ways of using the existing ones
- Structured resources for learning about proteins
- New experimental discoveries in structural biology
- New political aspects of bio- and information- technology
Also consultants will come (and possibly go as their careers IRL
change).
Styles of learning
This is dealt with elsewhere, but
the first thing that we shall be doing as consultants is throwing
ideas into the annealing pot. Many consultants have already been
involved in courses IRL and know what works and what doesn't.
Some hypothetical (perhaps extreme) ones might be:
- "I start by asking students to read 4 papers in Nature
about structure and report what they do and don't understand.
- "All the students build a physical model of myoglobin before
the course starts".
- "Students get 10 structures from
SWISS-3D images and have to
describe what each one might do".
I'm expecting that many ideas like this will come up and we'll
gradually anneal to which are most practical and valuable.
Students can also have an impact on how the course is taught.
You may have valuable experience in teaching related disciplines
or have helpful material (e.g. 'How to use xv').
up to index
pmr for pps
Sept 18 1994