2/12

Rawoot
Mason
Green
Sullivan

Good intro.

Problem 1. (See "Protein folding. Starting notes." ) Lattice folding.
Matrix representation. Good. Did not really have to go into details of
the algorithm at this stage.

2D lattice: great idea. Do you really need 3D? Probably not.

Problem 2 (required) almost not touched upon.

------------------------

Avent
Novitzky
Sharp

Good intro.

Problem 2.

Genetic algorithms (GA). Interesting thought. But "meshing" is a real issue.
Also, remember, your objective function is a fully differential, analytic
function. Why not use that? Test cases?

Problem 1. Do you really need something as complex as GA for the 2D lattice?
             Structure's energy is really easy to compute, see the Science
                        paper.

---------------------
Lehner
Giesy
Sanyal
MacQueen

Good intro.

Problem 2.  Good overview, but kind of off target. Remember, you will be
given an energy function, so this is not your concern. E(x,y,z) will be
given. You need to identify method to find min(E).
      Also, no pre-existing proteins. Again, see problem statement: no
similar proteins exit.

Problem 1   Not really touched upon.

------------------------------------------------

Dean,
Perich,
Abdelmegid,
Tsao

Good intro.

Make sure you are clear on the mathematical abstraction of the problem.
(minimization problem)

Did some reading: great!

Part 1
Interesting idea: break into parts. Divide and conquer. But do you
really need a complex algorithm for such as "simple" problem?


Part 2.

Somewhat off target and too generic. Remember, you will have a code, so no programming. The problem here is to figure out which of the existing numerical
methods might work.

-------------------------------

Kordell
Howell
Gaddis
Lee

Good intro.


Probem 2. Good general discussion, but you really do not need much of it --
the specific energy function is given.
         No need to discuss homology etc. -- irrelevant for our problem.
         Can jump into de-novo right away.

          Good use of graphs! Good grasp of Monte-Carlo

Problem 1. No specifics given.

-------------------------------------------------------------

Alemayehu
Williams
Brown
Keely

Good intro

Problem 1. Nice description.  Do you really need 3D? Good level of detail.

Thoughts on algorithm. Thoughts on mapping of real problem to the simplified
one. No need to complicate lattice. Rectangular is fine.


Problem 2. Lattice is irrelevant. Energy function is given. But then
one group member clarified.

Problem 1& 2 mixed, somewhat confusing.

---------------------------------------------------

Parker
Thomas
Evans
Ma


Good intro.

Problem 2. Interesting idea to try to divide and conquer. Generally,
this part is too generic. Your problem is more specific. Motionless
protein? What does it have to do with the problem?

Problem 1. Too generic. This problem is very specific.

----------------
Chen
Liu
Yohe
Janney


Good intro, but a bit too generic.


Problem 1 (Lattice).  Good thoughts, but still a bit too generic.

Problem 2. QR?  Which matrix do you mean? Which eigenvalues? How relevant to
the problem?   Too specific without explanation, will likely lose your audience

You can't talk about a complex, non-intuitive method without first explaining, 
at least briefly, what that is and how it works.