My last week as a DREU intern has come faster than expected, well I’m still going to work on my project so it’s not really the end yet lol. I continued working on the protein binding site project, cleaning and testing the code with my mentor and teammates. Also, my mentor, myself and one other teammate started a Tuesday, Wednesday, Thursday 15-minute 8 am scrum meeting to get ourselves set up for the day. Mondays and Fridays were left out since we have the bio subgroup meeting on Monday and the full group meeting on Friday. I also worked on the final DREU milestone requirements and finishing up my blogs although I might continue journaling the rest of the project for as long as I work on it. I am going to be a little busy next week but I am looking forward to some rest after that and then give my all to finish up this project.

I missed the 10 am meeting for this week and that is not recorded unfortunately. But I still got into my work for the week with high spirits. I created an XML file for the system testing the functionality of the Dynamic Regions Probabilistic Roadmap planning strategy. I run into a number of issues and the test still has not passed but it seems I am missing a file that hopefully someone in the lab would write so I can continue the project. I also spent a lot of time this week coming up with ideas and finally got my first draft of the presentation slides ready for the practice this Friday. The practice went really well and everyone came up with really great presentations and we also got great feedback so I’m looking forward to making those changes and see how great the presentation would be. I would try to include the video after I’m done with that in one of these blogs. This week has gone by surprisingly fast and I feel like I could have done more but we’ll keep that spirit for the following weeks.

I had my 10 am subgroup meeting and did some work on the PPL open source project: I started work on mapping the strategy following the pseudocode to implementation. Later in the week, my mentor took over the mapping so I could work on creating a system test for the code. Still, I was lucky to have a discussion with my mentor and other members of the team on the article that discusses the Dynamic Regions PRM strategy. I then cloned the branch from the lab’s gitlab server and moved the files I need for the test. At the Friday meeting, we discussed having a practice presentation for the REU showcase on the 26th so I have to get started on putting together slides for my presentation.

I mentioned last week how I had a slow week and had to deal with that, little did I know it was not going to be just one week. I am not sure whether it is because Monday was a holiday or because I was still not feeling my best or because my mentor was unusually busy but this week was probably the most unproductive for me. I was able to figure out the errors I was getting from my code, which it turns out was working all along but I was running the scripts in the wrong order: there was a script that run the other scripts in series but I wanted to check if one of the scripts I changed was working so I run it before the others but it would not work because it needed some files generated from the other scripts. I also resumed work on the Open-Source project but my mentor is taking over now so I would make unit test for those. I would get to work on those next week along one other assignment. I am glad I know what I am working on ahead of time so I can get to it even if my mentor is busy. I am still feeling a little unwell but I am hoping the last three weeks are very productive. Since the I already have images of the code I worked on, I am including the pictures below to make up for this week. And just to be clear, the pictures are for the same protein but covering the different things we look at.

Another week of internship is upon me, and I just realized that I am more than halfway through. We had the Monday meeting and oh one thing I forgot to mention that during last Friday meeting, the lab I am working with (Parasol Lab) announced that they are open sourcing their planning library and I get to work on that too with my grad mentor starting this week. In my previous blog (last week), I left myself some work to do but before I could get into that, I decided to run another test example for the protein binding cases; this protein has 9 binding sites. This caused an issue: the skeleton was being duplicated, one skeleton was being generated for all 9 binding sites but we only need one since that is enough to give a picture of the free space in the protein. So, I got to work. I have included a picture of how the skeleton looks like before and after I edited the code.

I started off week 5 with the usual 10 am sub group meeting where my grad mentor discussed the projects the members of the subgroup have been working on as well as ideas that could turn into new projects. I decided to try another of the protein/ligand binding site problems - with a different protein and ligand. That required me to make some changes in the test scripts which was cool cause I would be doing that when I get started on trying new test cases.

I have forgotten to include the meetings I attend every week but I would remember to from now on. I started off my fourth work week with a biology group meeting where we had updates on some ongoing or upcoming biology projects for the lab. Then a meeting on discussion of the module 2 of the pmpl & motion planning crash course. I moved to the crash course on the project I’m working on: the computational biology module. After which I went to the lab’s gitlab and cloned the repository for my project, mainly the binding sites branch, and tried out a given example - I executed code to produce a roadmap for a protein and it’s ligand. The cool thing about this planning is that there are different variables that can be considered in creating the roadmap: energy, clearance, etc and the roadmaps might or might not be the same as seen in pictures 1 and 3.

With a slow second week, I came into the third week with the aim to complete the goal I set for myself for the third week. I finally completed my first and second DREU milestones. For module 1, I was still left with the discussion of the Probabilistic Roadmap article so I could finish off my discussion questions. For the remainder of the second module, I made changes to my code to make a comparison, unlike last week, the comparison this time was between different k values (which based on my understanding of that line of code just describes how many nodes are connected to a new generated node). The k values were set at 1, 2, 4, 8, 16 and 32.

As I said in my previous blog post, I transitioned into week 2 with Labor Day holiday on the Monday and I moved on Tuesday. I went straight into the module 2 of the pmpl crash course which was a lot more hands-on than modules 0 and 1. Although my plan to complete the module did happen, I was still able to get through half of it and also had the article discussion which covered another paper (one of the three I had to read for module 1). The first section of module 2 required that I worked on a motion planning problem of my choice; since my project is about proteins which undergo folding, I decided to choose a box folding problem. Unfortunately, I couldn’t generate a roadmap for the problem, not even after trying all the samplers. The problem is that all the samplers are probabilistic roadmap strategies and so it was impossible for me to know whether the program would generate the roadmap and path if I let it run for a long time; I lost patience after 5 hours.

I started off my DREU experience with three welcome and introduction meetings spanning the first three days: a one-on-one meet with my grad mentor, a meet with the graduate students and fellow undergraduates working on computational biology projects and finally, a meeting with all DREU 2021 members and faculty. By the middle of the week, I obtained my Net ID and password which allows me remote access to the parasol lab’s computers (I am working with parasol lab). I then started my onboarding process, the zeroth module of the lab’s motion planning and probabilistic roadmaps crash course, by creating a page for myself on the lab’s website, setting up my gitlab, reading on what is expected of me for the summer. I was also added to the lab’s slack page as well as DREU’s slack and discord pages. That concluded the onboarding process and I moved on to the module 1 of the crash course which I had to read three articles, the third one on probabilistic roadmaps was a little difficult to understand but it was still fun. Thankfully, I am having a discussion on it with my grad mentor next week. The other requirement was to checkout the PMPL library (I ran the code and visualized it: I’m including a shot of how pretty it looks lol. We use vizmo for the visualization of the pmpl: Figure 1: showing the execution of a pmpl code in a 3D environment.