Week 1: May 26 - June 1
Well, this week was pretty much spent settling down. We had a lot of paperwork to take care of, regarding my room in the dorm, and my meal plan. And not to mention getting an account with the CS department. But all in all, it wasn't so bad.
Jenny and Glen picked me up from the airport on Saturday, and helped me move in. They're very nice, and I hope to see more of them.
It's very hot and humid here, a far cry from the fair weather I'm used to, but I guess it isn't as bad as it could be. My room isn't very nice though. It's very dark, because there's only this one tiny window, and there are millipedes everywhere. I requested a dorm change, to a cheaper and hopefully lighter room.
Dr. Amato is not here yet. She and Guang, the grad student who is also working on this project, are in China, giving a series of talks. In the meantime though, I've been given a bunch of papers to read, to get me caught up to where we are. I also got to meet Chris, the other undergrad who working on this. He showed me where all the files are located, and how to use the CVS file system.
I guess my main task for the next couple weeks are to orient myself.
(Here's a list of all the papers I read)
Week 2: June 2 - June 8
I got to meet Dr. Amato! She's a very interesting person, with lots of energy. I think I'll really enjoy working with her. We talked a bit about where we're at in the project, and where we'd like to go. We decided one of the first things we'd like to do is paralellize the code, since that could really speed up testing, and allow us to get results from larger proteins. As is, it takes at least overnight to run the program for moderately small sized proteins. So for the next week or so, I'll be learning about parallelism, and reading more papers.
Week 3: June 9 - June 15
This week, we are working to paralellize our code. Luckily, our code is based heavily on code written for general motion planning, for which a parallel version has already been written. After studying the differences between the sequential and parallel code, I feel it should be very easy to incorporate the changes necessary to make our version run in parallel.
Week 4: June 16 - June 22
Guang is now back from China. He offered insight on why we weren't successful in our earlier efforts to parallelize, and now we have a running version!
I ran some tests on this new code, at night on the dedicated queue, so that it wouldn't be interrupted by other programs. Unfortunately, it seems that for lots of processors, we get some kind of bus error. We think it might be something to do with the code we based ours off of, because in the parallel version of that, there are also some bugs that pop up occasionally.
In the meantime, we are also starting to work on some other ideas involving secondary structure formation, and how they relate to the folding pathway.
Week 5: June 23 - June 30
This week was a little hectic. Beth and I have been invited to participate in the USRG program, which is A&M's own undergraduate research program. On Thursday, we turned in abstracts, and gave short three minute presentations on our project. It was a pretty new experience for me, but I guess I'm glad I did it. I'm at least slightly less nervous about public speaking now.
So, I spent most of my time working on my presentation and abstract, and didn't have much time for actual work. I did start writing a couple functions for determining secondarty structure formation order.
Week 6: July 1 - July 7
Chris left for Europe, so we tried to get as much work done together as possible. We completed several functions. One is a new method for finding paths in our roadmap. Another determines exactly what secondary structures exist for any configuration. I'm not sure just how many of these we'll actually incorporate into our code, but I guess they're still useful to have.
I've also stopped working on parallelization, to concentrate exclusively on our analysis. I guess Dr. Amato feels this is more important, and something I can do with minimal supervision.
Week 7: July 8 - July 14
This week I began putting our new functions in the code. Most of them work now, or at least compile and don't cause segmentation faults. I also wrote another function for finding pairs of nodes between two paths. I'm quite proud of it. So far it seems to be working very well.
(a little later in the week)
I started testing our functions. So far the results are very encouraging - our new path selection method finds more paths, and ones with new secondary structure formation patterns. Next week, I'll start testing our path similarity functions. Hopefully it'll go as well as this week did.
Week 8: July 15 - July 21
More of the same from last week. I integrated our similarity functions, and they seem to be working alright, although some of the data we're getting back is not what we expected.
I've also started thinking about the presentation I'm making in a couple weeks.
Week 9: July 22 - July 28
Well, this week has been pretty much gearing up for the big presentation next Wednesday. We just found out we have to write a paper too -- I guess we knew all along, but I wasn't sure if that meant me too. But that's okay, it shouldn't be too difficult....
I feel like I should be wrapping things up, but it seems like there's something new to try every day. I'm told though that's just the way research goes sometimes.
Week 10: July 29 - August 4
(writing this from San Diego)
Wow! What a long week that was.
We had our presentations on Wednesday. It went well, better than the practices we gave on Monday, I think. I was less nervous, which made all the difference (to me anyways). Then afterwards, I got to rest a bit, but not too long, because there was still that big paper looming over my head. We got a little bit of an extension on it, due to the circumstances.
And somewhere between writing this paper and closing up business at the University, I found time to pack and otherwise get ready for coming back.
All in all, I'd say it was definetly worth it. I got to see my "hometown" of Austin, TX, which I hadn't seen since I was a toddler. I had a good time and some good laughs, learned a lot, and met some really great people. In general I feel more confident, a little more secure with my future.
If you're reading this as a prospective DMP participant, do it! It's probably one of the best things I've done. On that note, I'd like to thank the CRA and Drs. Amato and Welch for this opportunity. And thanks to everyone else who made my summer in Texas as enjoyable as possible.