This was my first year working under a research fellowship, and second in research in general. The REU aided greatly financially and left my worries to the research alone. I also appreciated the liberty afforded by the fellowship, and was able to continue a project I had started the year before with another professor. My work involved theoretical computer simulations of protein folding. Specifically, I had to verify the model we were using by comparing it to experiment, in order to prepare for further simulations modeling the aggregation of beta-amyloid fragments that are linked to the onset of Alzheimer?s disease. My daily tasks included writing programs, making graphs and tables of the results, and writing and editing the paper about the results I intend to submit to journals. The REU enabled me to experience every aspect of research. As described above, I was involved in every step of the research process, all the way through the writing of the final paper. Our weekly meetings allowed me to share the experience with other undergraduates and not feel overwhelmed by all that I was learning and doing. Once the paper writing process got underway, the weekly REU meetings were especially helpful due to the time we spent on peer-editing and feedback. Finally, the ethics meetings were interesting and encouraged critical thinking about ambiguous subjects. Unfortunately, the students in our group all started the REU at different times, so it was not clear when mine actually started and ended. For the same reason, we did not have a final meeting to conclude the experience and also did not get to go on the promised field trip to the Field Museum. Furthermore, the ethics meetings lost steam and were not very enlightening towards the end. But despite these few things, the REU overall was a very educational and eye-opening experience for which I am grateful.

This past week (ended 11th August) I:

Spent most of the time in San Francisco and didn't do a whole lot of work. I did finish plotting all data for the explicit trajectory and worked on my manuscript. This week is entirely devoted to writing.

This past week (ended 4th August) I:

Finally finished and validated the correlation function program. I put all the newly generated data into my manuscript. We also came up with more trajectories I should run to supplement the results I put in my paper.

This past week (ended 28th July) I:

Ran my job on Teragrid and found out it takes 2hours per nanosecond of trajectory. I will submit my other 1+1 jobs this week. Last week I also implemented various programs to get rid of translation and rotation for the correlation function, but none performed better than any other. I went back to the correlation code and found a bug and in fixing that I immediately got better values. However, it is still unclear if they are accurate or just close to what is expected. This week I am writing a new correlation program that focuses on the values at infinity, ie the order parameter. I have learned not to get ahead of myself and overlook potential problems that will only get bigger with time.

This past week (ended 21st July) I:

Found out the correlation function needed to be computed not just rotating about principal axes, but also minimizing RMSD. So I had to scrap all my work from before and use a new program that did the above. However, in watching a video of the trajectory, there appears to still be alot of overall motion, so perhaps the program is working properly. I also waited in the queue for a long time on teragrid and finally my job was run, but there was a bug so I will have to redo it this week.

This past week (ended 14th July) I:

Began plotting the correlation function for all my force fields at various time segments. I had trouble verifying my results, but performed several tests on the program and got predicted output so think everything is working. I also ran my first job on Teragrid to see how much I can get done per unit time. This will help me anticipate the length of the trajectories I will be running this week.

This week I:

wrote a program to compute the correlation function for my protein and prepared the 1+1 simulations.

Next week I will:

run the 1+1 simulations, plot the correlation function to get the S-squared order parameter, and add the results to my manuscript.

-- AndrewKent - 12 Jul 2006