Bioinformatics Seminar (MTAT.03.242)

Course info

• 1st seminar (overview of the course) 25.02
• Seminars: Monday 10:15 - 12:00, J. Liivi 2 - 122
• Contacts: Ahto Salumets (ahto.salumets [at] ut.ee), Kaur Alasoo (kaur.alasoo [at] ut.ee), room 311 in J. Liivi 2
• BIIT (Bioinformatics, Algorithmics, and Data Mining group) group website : http://biit.cs.ut.ee/

Bioinformatics is an interdisciplinary field that combines computer science, biology, statistics and mathematics for understanding biological data.

Learning objectives

• Learn to critically read and evaluate state-of-the-art bioinformatics research literature.
• Learn to compare competing computational approach to answer the same biological question. Identify the strengths and weaknesses of different approaches and the trade-offs involved.
• The seminar is meant to complement the Bioinformatics course thought in parallel. Students without previous background in bioinformatics are strongly recommended to also take the course.

Aims of the course:

The broad aim of the seminar is to provide basic introduction into the biological questions, introduce common measurement techniques and data types and demonstrate how combining rich data with clever computational approaches can lead to new scientific insight. This semester's seminar is complementary to the Bioinformatics course and topics will be chosen to roughly match the course schedule.

The seminar is open for students from all backgrounds. However, for those familiar with the field, here are some keywords that might (or might not) be covered during the semester:

Experimental: transcription factors, enhancers, gene expression, chromatin accessibility, ChIP-seq, ATAC-seq.

Computational: clustering, principal component analysis, k-mers, missing data, differential expression, deep learning.

As a bonus you will get a great experience in functional reading of a scientific paper and improve your presentation and teaching skills.

Seminar organisations

1. We will have one seminar in every week. The date of the first seminar is TBD.
2. In each seminar, two students will make two short (15-20 minutes) presentations about two related scientific papers (see Topics). These could be, for example, two different computational approaches to solve a particular biological problem.
3. The presentations will be followed by discussion, with the aim of comparing and contrasting the two different papers in order to understand their relative strengths and weaknesses. To aid the discussion, 2 students (reviewers) will fulfil specific roles.
4. For the next sessions, presenters will write a report (400-500 words), comparing the two methods and summarising the discussion. Each report will be reviewed by two students who grade them.
5. We expect every student to be familiar with the papers presented in each seminar.

Roles for reviewers

• One of the reviewers will lead discussion (or they can share this role). This means introducing the papers, moderating the discussion and making sure that everyone will get a word.
• Skeptic - asks at least 3 questions that reveal potential shortcomings of the methods.
• Optimist - asks at least 3 questions about potential applications of the method or potential future directions.

Others will get points for each questions that was asked.

Instructions for the written review

Each reviewer has to read and provide feedback to reports written by both presenters. While reviewing, focus on the following aspects of the work and rate each of them on the scale from 1 to 5:

1. Is the biological problem that the two methods try to solve stated clearly?
2. Is it clear, what is the input data to the methods and what do they produce as output?
3. Are the main differences between the methods stated clearly?
4. Is there a clear conclusion in the end? Is one method always better than the other or are the two methods better in different niches? Justify your decision!

In addition to assessing these points, you should also provide clear and actionable feedback to the presenters. Avoid simple statements such as 'vague', 'unclear', 'hard to follow'. Instead, give explicit suggestions of what should be done differently or write down explicit full sentence questions.

Requirements to pass the course:

The seminar is a non-differentiated course (pass/fail). You need to collect 70 points out of 100 in order to pass the course.

1. Presenting an article/topic gives maximum 30 points. The presentation points are scaled according to listener feedback which evaluates both clearness and presentation skills.
2. Writing a report 20 points (20*average points from two reviewers)
3. Being a reviewer (also leading discussion) gives 20 points
4. Asking questions 1 point per question (up to 5 per seminar, needs to be meaningful!)
5. Attending all the seminars gives 20 points (necessary to participate at least 70% of seminars!)

Current state of the points can be seen here