JPB1071H - Advanced Topics: Computational Neuroscience

Course Coordinators:

Description:

Computational Neuroscience (CNS) seeks to understand the fundamental principles of neural dynamics and how the brain and nervous systems compute.
It combines experiments and theory and encompasses multiple disciplines, including but not limited to Physiology, Physics, Mathematics, and Engineering. This course aims to provide an essential overview of fundamental concepts in CNS, introducing some of the current theories, models, and methods used in the field. The course is strongly interdisciplinary and offers self-contained lectures spanning different aspects of CNS, from fundamental single-neuron and neural network models to modeling in Neurosurgery and Biomedicine.

The course’s Syllabus is finalized every year around November. Topics presented may vary on a year-by-year basis depending on the availability of each lecturer, but are generally in the following list (* denotes topics presented regularly):
● Biophysical models of neurons*
● Synaptic models*
● Models of Cortical Networks*
● Neural Encoding*
● Synaptic plasticity
● Learning
● Bayesian Models in Psychiatry
● Behavioral Models
● Whole Brain Modelling
● Mechanisms of Brain Rhythms
● Memory Paradigms and Hippocampal Networks
● Biophysics of Deep Brain Stimulation
● NeuroAI and closed-loop neuroscience*

Format:

This graduate-only course is presented in seminar-style, self-contained weekly lectures. The course consists of 10–12 lectures, followed by 2–4 classes for students’ presentations; the exact numbers depend on the lecturers’ availability and the number of students enrolled. Students are asked to present a peer-reviewed article in computational neuroscience from a list provided beforehand, during the course. The final written assignment is a critical report on the same article.

Pre-enrolment is required, and students who are not from the Department of Physiology will be contacted by the Department and asked to reach out to the instructors for pre-approval to enrol. Attendance in the course is limited to enrolled students only; auditing is not accepted.

The field of Computational Neuroscience is represented by annual meetings, journals specific to the field and is recognized via symposia, socials and topics at the annual Society for Neuroscience meeting. Physiological journals strongly encourage modeling and quantification. Almost 20 years ago, the chief editor of the Journal of Neurophysiology claimed that, "...Many of the most important findings in neurophysiology come from the use of quantitative methods of data analysis and from models of nervous system structure and function. Therefore we invite computational and theoretical papers that are strongly tied to the physiological analysis of the brain and nervous system." (J. Neurophysiol. 88:1, 2002).

The Department of Physiology has a history of theoretical physiology and encourages students with backgrounds in physical sciences. This course is a natural addition to the graduate program given the developments in the Neuroscience field today.

Prerequisite:

A solid foundation in Calculus (at least at the first-year level) is highly recommended, along with an introductory-level knowledge of biology or neuroscience. No programming skills are necessary.

Evaluation:

Oral presentations: 40%
Written Final Assignment: 40%
Class Participation: 20%

Remarks:

Maximum enrollment: 20

Course Location: Please note that this course takes place off-campus at the Krembil
Research Institute (60 Leonard Avenue, Toronto, ON M5T 0S8)
Course Time: Thursdays 1:30-3:30 PM

~ Last updated: 3-Nov-2020