Los Alamos High School (68)/Interim Report

Interim Report
http://mode.lanl.k12.nm.us/get_interim1112.php?team_id=68

Problem Definition
The aim of my project is to simulate the most effective way in which synthetic proteins can fold. Protein folding is a very important problem in biology, because it is involved in nearly all living processes. The failure to fold is associated with many well-known diseases. In order to find effective folding sequences, I plan to use the open-source Protein Database and Python-based simulations to explore this problem. I will use a genetic algorithm to find and design artificial proteins that fold extremely effectively. I hope that this project will yield some interesting insight to protein folding. If initially successful, I will investigate GPU-based algorithms and specialized Python libraries to accelerate my program.

Problem Solution
The protein will fold in a controlled environment, one similar to a cell interior. The genetic algorithm would start with a completely randomized set of proteins, fold them, and then eliminate all those except for the top 10 percent. It would then repeat the experiment with the "children" of those top 10%. These "children" are formed from a pool of the "parent's" "chromosomes." This process would be repeated thousands of times.

I plan to utilize a multiple stage solution in order to develop a better genetic algorithm. I will initially use a basic genetic algorithm to solve a simple problem such as determining the minimum of a complex function such as Rastrigin's Function. After creating the foundation for my final algorithm, I will design a genetic algorithm to optimize the design of a complex dynamical process. One application of this would be to optimize a process such as designing a vehicle. Finally, I will use a complex genetic algorithm to determine the best folding proteins that fulfill certain parameters. If necessary, I might have to include certain chaperone proteins, or use a method that acts like a chaperone protein.

Progress to Date
I have finished research on proteins, genetic algorithms, and determined my method for programming this. I have also learned several new things within python, such as classes, dictionaries, and additional list operations. I am also nearly done with programming my initial genetic algorithm, but it is not ready for debugging yet. The project is moving ahead smoothly.

Expected Results
A python-based program with an implementation of a genetic algorithm to solve multiple problems Using the above program to find the minimum of a higher-dimensional function Optimal design of a dynamical process, using the above program Using the algorithm to design optimal protein folding in two dimensions If there is time explore protein folding in three dimensions

Problem Statement
In my opinion (Cesar Venegas, from NM Tech) this is very well though out problem with a very focused plan that can yield interesting results the report clearly states a problem that will seems to have legitimize solution. Its nice that you have researched the characteristics of the proteins that you are studying this will prove to be beneficial within your work.

[4/4]

Mathematical Method
Your proposal seems very well organized as far as programming goes. Although there is no concrete evidence as to how you will approach this problem using a mathematical method. Although you have an idea of what you will be doing in the program in which you will only take the 10% of each protein is a good start to explaining the mathematical method.

[3/4]

Computational Method
Your computational method seems very complex as far as the programming goes but then again this is a very complex problem my only advice is try not to burden yourself with to much work to where you will not be able to finish the programming. If you can successfully program using python I think that will be impressive.

[4/4]

Results and Conclusions
The results seem like they will be very complex. In fact in your proposal you only mention what the out come of using the results will be. You do not state what the benefit of the final results will compare to you expected. I would take a little more time to re-evaluate your results.

[3/4]

Code
I think your code is very original and very complex. As I mentioned before the only problem that can arise from this project is having such a complex code that you in fact will have problems trying to debug it and finish it before the deadline. The fact that you have begun working on the code and have a focus as to where you want to go with your project is great.

[4/4]