*grabs you personally by the throat* SAVE YOUR SOUL. DO NOT TAKE THIS COURSE.

Unless you want to do 4 projects that have very little guidance in the specs, each split into 2 parts that require different types of outputs, on top of doing 7 homeworks, each with 4-10 Leetcode-like coding problems evaluated on a shitty $80 online textbook website (Stepik) where you have to download the input to your computer, run the code, and upload the outputs and pray that you matched the formatting exactly, otherwise repeat the process. (Also, I get Stepik advertising emails in Russian, which 1) I did not sign up for 2) I don't know Russian.)

And about those projects, they decided to try something new this quarter: making us upload our results to a bioinformatics leaderboard website. The fun thing is that 1) They don't post the leaderboard until 2-3 days before the project is due. 2) Someone has to manually approve that you can join the leaderboard. Which means you wait for some poor TA to handle your request. 3) They don't post the grading thresholds WITH the project spec or even when they post the leaderboards sometimes, so if you finish your code early, you have to wait for the announcement of the threshold. If you don't pass it? Guess you're working on the project again!

I have never taken a class with this many Canvas announcements about project extensions and grading thresholds and about homework problems becoming Extra Credit because very few people are successfully solving it. And have I mentioned that they also made us read 4 papers and ask and answer other students' questions about it? To me, it felt like the blind leading the blind.

The only saving grace of this course is that the midterm was reasonable. If you read and understand the textbook and the slides (which is what I did because their lecturing is Pretty Bad, especially Ernst's), you can do the problems. They just make you apply the techniques to the given data. They also gave a set of practice problems that matched pretty closely. I'm writing this review before the final though, so maybe they decide to completely switch it up on us. (But, I'm skimming the final practice problems, it seems like it's the same problem format.)

Who knows. Maybe you'll enjoy the torture more than I did. Maybe you're that kid who was already working on a bioinformatics library for their research and used it for Project 1, landing you a score in the top 3, at which point you're obligated to do a presentation of your solution to the class. The class is mostly empty, by the way. Just like how this class made me feel.

Grade breakdown for Spring 2023: Projects 25%. Homeworks 20%. Midterm Exam 25%. Final Exam 25%. Paper/Guest Speaker Question and Responses 5%.

First, even though STATS 100A is a prereq for this class, it isn't too stats heavy, so if you want to take this without having taken STATS 100A, talk to the TAs or the prof and they'll probably let you in.
This class is an interesting application of algorithms in biology, and largely was divided into the following:
1. Sequence Alignment to the Genome (Leetcode string problems, but on large strings)
2. Assembly (graph traversal problems/ path finding problems)
3. RNA Sequencing (more string problems, dynamic programming)
4. Hidden Markov Models

I found the material to be pretty interesting, and since I took it along with CS180, there was some overlap between the classes as well. Eskin himself is a super smart guy, but isn't the best lecturer. He often went into detail on advanced topics that weren't important, and didn't have slides for RNA Sequencing even though it wasn't in the textbook. However, the TAs for the course were super helpful and reviewed the material which was super helpful.

The class involved reading small research papers, programming homework problems from the textbook, programming projects, a midterm, and a final. Of those, the homework problems and projects were the most time consuming.

The quarter I took 122 was an experimental quarter: Eskin wants to make the class more difficult since (in his words) "people are doing too well in it". However, for the first 3 projects, the starter code provided by the TAs was good enough for full credit, which was a bit of a joke. Expect this to change in future quarters. One of these projects, sequence alignment on a 100 million length genome was so hard and stressful that they had to make it extra credit. Also, this was the first time they ever had RNA Sequencing projects, so there were some teething troubles with it, but this should get easy in future quarters.

On all projects, the TAs made it seem like they were easier than they actually were (assuming you did it diligently). It helps to start early on the projects and ask the TAs for help.

The exams were fair and easy. They gave out a practice exam which we had to solve and turn in, and the real exams were similar to the practice ones, which was really helpful. The final was non-cumulative as well.

If you're a CS major, you may want to do a little review of biology before you take this class. They jumped straight into the material, and as a result, I was lost for the first couple of weeks. Also, this is a great class if you're interested in bioinformatics research. Talk to him in office hours or after class and if you show some interest, he'll probably take you in.

Overall, it's a pretty chill class apart from 2 of the projects. I'd recommend taking it. The grading is pretty good, too.

The bioinformatics classes seem to change a lot from quarter to quarter, but here is Spring 2023.

Pros:
- Lectures are not necessary to do well.
- Lectures, tests, homework, projects, and papers all synergize together extremely well. You can watch a lecture about an algorithm, read the paper they give you that describes it, work out individual subroutines on the homework, and put it together on the project. I didn't figure that out until later, but it's pretty special.
- You feel like you learned something hard and useful.
- Opportunities for extra credit or easy points.
- Very easy tests.

Cons:
- A constant barrage of work from beginning to end. The sheer volume and difficulty of the homework and projects is incredible. For reference, the final project of CS CM121 was to align sequencing reads to a small genome and find what genotype it had at a specific point in the genome. Since the genome wasn't that long, you were just supposed to brute-force align a bunch of reads and use Bayes Theorem to find the most likely genotype. For CM122, the FIRST project of 4 or 5 projects was to do the same, but with a genome millions to a billion bases long with mutations. You have to employ a very interesting fast substring search algorithm to do that and figure out a way to account for frameshift mutations. I learned to tell my Mac to stay awake while it was closed so I could run the algorithms overnight.
- Feedback is hard to come by, especially for the homework. Each homework question asks you to write a function and test it with some sample input before running it with the question's input. I can't count how many times my code worked with the sample but not the graded input and I had no idea why.

I think they should get rid of 121 and spread this class over the two quarters.

This class seems really bad until you realize that the exams are almost one to one copies of the practice exams (at least they were for spring '23). I glanced over the practice midterm and to my surprise when I took the actual midterm it was basically the exact same with different numbers. From then on I did not go to a single lecture, partly because they were really bad and hard to follow. All I did was study the practice final and ended up getting a 100 on the actual one. The issue is that the workload in this class is actually insane: weekly homework + weekly projects + discussion posts. It does however become manageable once you realize you don't have to go to lecture. I would recommend going to discussion and office hours for help on the projects and homework, particularly because the platform the homework was conducted on was atrocious. They also did some unreal grade rounding at the end. Based on the grade break down in the rubric I calculated I would get around a D- but somehow ended up with a B-. Considering I barely did discussion posts and only completed a little more than half the homework and projects I'd say this class is pretty easy to get a good grade.

I hated this class. I can't say too much about the grading and the exams because I dropped the course before the first midterm. But if you are just a CS(-related) major without any interest or background in biology, be VERY careful taking this course. I personally had a hard time especially since they reworked the course. As I heard, the projects used to be guided with template code to work with, but Spring 2023 did not receive ANY of that. The Stepik assignments were extremely challenging and draining, especially if you aren't experienced in Leetcode, and you are completely on your own figuring out how to do the projects. I literally could not figure it out and the TAs I asked were so vague so I dropped before even finishing Project 1a.

Strangely, they recorded every class but did not post them. You had to have a legitimate reason to get the recordings and you'd email them for it.

Don't let the grading fool you into thinking this is a grade-safe elective, either. The class is cross-listed for graduates as well so I'm assuming that's where the higher portion of grades come from.

There is very little hand-holding and you need to figure out a lot on your own, especially if you are just coming from CS 32 and that's all the programming experience you've had. Unless you are in love with bioinformatics and would be okay dedicating all your free time to self-learning this course from the ground up, please DO NOT TAKE THIS CLASS!

Took this class during COVID; the prof and TAs had some weird communication issues where the professor would say something about a due date and the TAs would lag behind a little on getting the news out to all students. They did their best though! Whenever more than a few people asked for an extension, it was given globally to the whole class. Eleazar kept calling these “revolts” jokingly, but they totally were. I don’t think we had a project due until week 6.

All things considered, it was a good class. You learn about genome assembly, resequencing, RNA sequencing in the context of computation. The Stepik homeworks were way more CS heavy than the projects, honestly. They offered a ton of extra credit and lots of cool guest speakers if you’re into research. The projects aren’t that hard and the exams are just there to see if you’re paying attention. If you’re into research/bioinformatics, I’d suggest this as an introductory course. Not too difficult and pretty cool. This is coming from someone who learned Python concurrently with taking this course!