BME 4550

When I came into this class, I didn't really know what Data Science was, but when I left, I felt like I had learned so much and could actually apply what I learned to research. In this class, you use Rivanna which is the Research Computing center at UVA and UNIX commands to go through a RNAseq workflow and use R for data analysis. Here's the breakdown:
ASSIGNMENTS:
- You have a weekly homework assignment that you can complete by looking at the lecture notes. You also get some time in class to work on the assignments and Civelek is really helpful in office hours. Also, protip is to really utilize the Piazza board! (He really loves seeing Piazza posts and in my particular semester, we had a lot of engagement and on one particular hw assignment, we all struggled together but got through it! :) )
EXAMS/QUIZZES:
- There are weekly quizzes, but he drops the lowest one.
- There were no exams!
OVERALL:
I would definitely recommend this class because you work with actual data from Civelek's lab and can see *actual results*. During the semester, you have an independent research project with a partner and you can choose what you'd like to work on as long as it's not RNAseq. (Make sure you start early so you're not scrambling at the last minute and there's also check ins) You also can get extra credit for attending BME seminars and Professor Civelek is absolutely the best because he really cares about his students and really wants to make the material as understandable as possible!

SIDENOTE:
Sometimes using Rivanna can get frustrating because a lot of research teams use the same cores that we use in class, but if you start the hw assignments early, you'll be fine! Also, make sure to double check your SLURM script before you submit a job for like 8 hours
#tCFspring2021

Do not take this course lol. Although you get a pretty good grade at the end, it is not worth your time or effort. If you want to learn real data science methods then take machine learning in the computer science department or a data science course in the data science school. You spend most of your time in the class learning basic Unix commands. You do not learn any data science techniques until halfway through the semester. As a result, they are all just rushed. All the important math is also skipped over (for most topics, it was there for PCA, but the post doc who taught that portion of the course did a pretty terrible job). The second half of the semester is also in R, but you also don't learn R until the middle of the semester. You are expected to make complex R scripts with little to no preparation. Give the students a couple more weeks and they will adapt to the language. If you do not know R, then do not bother taking this course. There is a semester long project that is very poorly structured. You are expected way too much out of it and it is very time consuming. The professor teaches you RNAseq bioinformatics tools but then says your project must be from any data set except for RNAseq. Although the methods are little similar for other sequencing data sets, way too much time is wasted learning about them. And you only gloss the surface so your end project product is low quality in terms of the field. Dr. Civelek doesn't really help you that much with the specifics of your project either since everyone's projects are so different. If you are interested in this course, wait a couple semesters until it is taught by a different professor or completely re-done. Learn data science in a different department or in a BME lab. You won't learn anything useful in this class.

Absolutely not. We spent the first month doing problems out of an Introduction to Data Science with R textbook with no relevance to bioinformatics whatsoever. Immediately following that month, he asked us to get into groups and propose projects -- never mind that some of us had no idea what, or how to conduct, any of the analyses the other review mentioned (Chip-Seq, Annotating a personal genome, Microbiome analysis, etc.) other than Civelek telling us to just look it up online. From the time I spent there, the only people that really succeeded were the ones that already had experience.

This was Mete's first semester teaching BME Data Science and it was kinda a 'guinea pig' experience overall. Structure-wise, the class consisted of one big group project on a specific topic (Chip-Seq, Annotating a personal genome, Microbiome analysis, etc.), a couple easy, some pre-class assignments, and a couple sporadic quizzes. It was probably my favorite BME elective and I would recommend the class to anyone with any interest in biological data/bioinformatics for a couple reasons:

1) There are not many undergraduate programs where you get any experience in bioinformatics. Mete does a great job with an introduction to FASTA formats, Unix, and R Studio. It's not too difficult or too complex, but it's great basic skills that would give you a good base if you wanted to learn more about it.

2) Mete is great himself. One of the most personable professors in the BME department. Incredibly nice, will learn your name in a couple classes, is always up for answering questions, and is flexible with due dates if you have a legitimate reason. If you show interest, he will give you as much help as you need.

3. I'm terrible at Java, was not a fan of Matlab, and loved this class! Mete was able to help the class find resources to help us out. The material itself really isn't that difficult. It's completely different from all the other BME classes, so it'll be hard due to its 'newness,' but with Google and your classmates on Piazza, it's doable.