Open Source: Computational Neuroscience Course

In the fall semester of 2020, the team Laboratory of Neurobiology and Developmental Physiology read a course “Computational Neurosciences” for students of partner master’s degrees at HSE and ITMO, as well as for interested audiences. The course has been held as part of the JetBrains educational programs since 2019. This year, unlike the past, the training format was, of course, distance – lectures and seminars were conducted in the form of videoconferences. During the course, students were offered basic material for study and discussion in the classroom, materials for independent, deeper immersion, interesting practical tasks on modeling neurons and biological neural networks.

The goal of the course is to give students an idea of ​​what and in what ways can be modeled in neuroscience and give them the opportunity to practice a little in this on a few relatively simple tasks. Prerequisites for a full-fledged assimilation of the material are the ability to program and an interest in biology, however, even if you do not know how to program, but are interested in issues related to the work of the nervous system and its modeling – you will still be interested in listening to these lectures!

The first part of the course covers key topics from neurobiology that students need to understand to one degree or another in order to try to model something: it tells about the structure and functioning of the nervous system at the organismal, tissue, cellular and molecular levels, about biophysical phenomena, the underlying processes that occur when signals are generated and transmitted, etc. Models of the same biophysical phenomena are also considered, as well as models of generation and conduction of action potential at the level of single and at the level of many neurons. Another, somewhat separate and slightly closer to medicine, chapter in this part of the course is electroencephalography (EEG) data processing. In parallel with the lectures of the first part, students are offered practical tasks – implementation of the classical Hodgkin model – Huxley, describing the characteristics of excitable cells, working with real EEG data and creating a model of a chemical synapse.

In the second part of the course, we talk about how information is encoded and decoded in the nervous system, what plasticity is and what causes it, how spiking neural networks differ from classical ANNs and how they can be used in biological research, how biochemical regulation of neurons is carried out, and how the nervous system develops from a fertilized egg to an adult organism. In the final lecture, we talk about how machine learning, artificial intelligence and neuroscience are related, what processes and phenomena in biology have served as inspiration for the development of new approaches in machine learning, and how machine learning is used in brain research.

If you are interested in our course or any individual topics that we touched upon, we invite you in the next fall semester: it will be even more interesting, since we are constantly expanding and refining the content. Anyone can attend the course. You can also watch all the lectures in 2020 at any time convenient for you – video materials are available at YouTube channel JetBrains Research.

Slide from the lecture on encoding and decoding information in the neural network.
Slide from the lecture on encoding and decoding information in the neural network.

Similar Posts

Leave a Reply