Sessions from the INCF Neuroinformatics Assembly 2022 Day 3. 

VIEW THE PROGRAM

This is a freely available online course on neuroscience for people with a machine learning background. The aim is to bring together these two fields that have a shared goal in understanding intelligent processes. Rather than pushing for “neuroscience-inspired” ideas in machine learning, the idea is to broaden the conceptions of both fields to incorporate elements of the other in the hope that this will lead to new, creative thinking.

This couse is the opening module for the University of Toronto's Krembil Centre for Neuroinformatics' virtual learning series Solving Problems in Mental Health Using Multi-Scale Computational Neuroscience. Lessons in this course introduce participants to the study of brain disorders, starting from elemental units like genes and neurons, eventually building up to whole-brain modelling and global activity patterns.

This course is currently under construction but will coming soon.  It will give an overview of the world of scientific publishing, spanning from traditional formats, to open to access, to open, interactive, reproducible, and 'living' publications with modifiable and executable code.

In this short series of lectures, participants will take a look at articles using TVB in a clinical context. Specifically, participants will see how TVB can help to predict recovery after stroke and how individual epileptic seizures are simulated. The course lecturers will briefly describe the methods used and results achieved in the articles.

This course, consisting of one lecture and two workshops, is presented by the Computational Genomics Lab at the Centre for Addiction and Mental Health and University of Toronto. The lecture deals with single-cell and bulk level transciptomics, while the two hands-on workshops introduce users to transcriptomic data types (e.g., RNAseq) and how to perform analyses in specific use cases (e.g., cellular changes in major depression). 

This course consists of 12 lectures on the visual system and neural coding produced by the Allen Institute for Brain Science. The lectures cover broad neurophysiological concepts such as information theory and the mammalian visual system, as well as more specific topics such as cell types and their functions in the mammalian retina. 

The dimensionality and size of datasets in many fields of neuroscience research require massively parallel computing power.  Fortunately, the maturity and accessibility of virtualization technologies has made it feasible to run the same analysis environments on platforms ranging from single laptop computers up to high-performance computing networks.

EEGLAB is an interactive MATLAB toolbox for processing continuous and event-related EEG, MEG, and other electrophysiological data incorporating independent component analysis (ICA), time/frequency analysis, artifact rejection, event-related statistics, and several useful modes of visualization of the averaged and single-trial data.

This course consists of two introductory lectures on different aspects of statistical models, in which you will learn about the neural coding problem, aspects of neural activity carry information, multiple spike train models, latent variable models, and regularization. 

This course includes both lectures and tutorials around the management and analysis of genomic data in clinical research and care. Participants are led through the basics of genome-wide association studies (GWAS), genotypes, and polygenic risk scores, as well as novel concepts and tools for more sophisticated consideration of population stratification in GWAS.

As technological improvements continue to facilitate innovations in the mental health space, researchers and clinicians are faced with novel opportunities and challenges regarding study design, diagnoses, treatments, and follow-up care. This course includes a lecture outlining these new developments, as well as a workshop which introduces users to Synapse, an open-source platform for collaborative data analysis. 

This course contains sessions from the second day of INCF's Neuroinformatics Assembly 2022.

The human mind is a complex system that produces, processes, and transmits information in an incomparable manner. Human thoughts and actions depend profoundly on the proper function of neurons. If this function is disrupted, degeneration and disease can be the consequence. This course provides insights into state-of-the-art views on neurodegenerative, neuropsychiatric, and neuroimmunological disorders as well as clinical neuroanatomy and clinical aspects of brain imaging.

Data science relies on several important aspects of mathematics. In this course, you'll learn what forms of mathematics are most useful for data science, and see some worked examples of how math can solve important data science problems.

There is a broad consensus among researchers, publishers, and funding bodies that open sharing of data is needed to address major reproducibility and transparency challenges that currently exist in all scientific disciplines. In addition to potentially increasing the utilization of shared data through re-analysis and integration with other data, data sharing is beneficial for individual researchers through data citation and increased exposure of research.

This course includes both lectures and tutorials around the management and analysis of genomic data in clinical research and care. Participants are led through the basics of genome-wide association studies (GWAS), genotypes, and polygenic risk scores, as well as novel concepts and tools for more sophisticated consideration of population stratification in GWAS.

This course tackles the issue of maintaining ethical research and healthcare practices in the age of increasingly powerful technological tools like machine learning and artificial intelligence. While there is great potential for innovation and improvement in the clinical space thanks to AI development, lecturers in this course advocate for a greater emphasis on human-centric care, calling for algorithm design which takes the full intersectionality of individuals into account.

As research methods and experimental technologies become ever more sophisticated, the amount of health-related data per individual which has become accessible is vast, giving rise to a corresponding need for cross-domain data integration, whole-person modelling, and improved precision medicine. This course provides lessons describing state of the art methods and repositories, as well as a tutorial on computational methods for data integration. 

Difficulties experienced in understanding machine learning techniques often stem from lack of clarity concerning more basic statistical models and fundamental considerations, including the various regression models that can all be subsumed under the General Linear Model.