Machine Learning in Structural Biology

Structural Biology is a rapidly growing scientific field that informs other areas of biology, as well as medicine and pharmacology. Machine learning-based approaches have been increasingly making inroads into structural biology. Machine learning is used for experimental data processing (NMR, XRS, cryogenic microscopy), generating new formulas, docking result analysis, prediction of protein-to-protein interactions, and prediction of the properties of chemical agents and their toxicity for humans. Machine learning is furthermore employed for the purpose of accelerating the traditional approaches in structural biology, such as molecular dynamics or quantum-mechanical modeling.

The central purpose of Machine Learning in Structural Biology is to introduce students to both the classic methods of structural biology and their twins, underpinned by machine learning.

The course lasts 1 semester, with 6 theoretical and 6 practical classes.

The course is geared towards Master's degree holders and graduate researchers in various fields of biology and chemistry. The course introduces students to the theoretical fundamentals of classic and ML-enabled approaches in structural biology, after which the students apply this knowledge in practical work and compare the results.

As a result of taking the course, the student will be able to independently formulate problems in the language of structural biology and solve them using state-of-the-art methodology.

msu.vsb.ai@gmail.com

We plan to admit 1 study group of 20 students.

Entry requirements:

• Competent use of Python and some ML libraries;
• A course in molecular biology, classical physics, organic chemistry.