STKD6810 Neuro-insights: Data Science Approaches in Neuroscience II Emneplan

Because most young researchers in life and health sciences do not have a solid quantitative background, they face difficulties when analyzing data independently. This difficulty represents a major drawback in research. Students waste time learning analytical methods by themselves that could be more quickly learned with proper instruction and support. Additionally, the lack of convention or standards in some fields is a source of confusion that slows the learning process. As a consequence, the quality of insights and research productivity suffer. This course provides a comprehensive introduction to data science and big data applied to neuroscience research.

Its content is designed to train the participants in state-of-the-art techniques in data analysis and machine learning. This will enable the students to interact independently with the data and draw insights from them. The modules are organized so the participants have the opportunity to learn how to handle the most common data types (e.g., EEG, calcium imaging). Special attention is given to field-tested data management protocols, as they are critical for a fast transition from data acquisition to knowledge generation.

This is a hands-on course where the students will learn from implementing the analysis themselves with close supervision. The course will focus on case studies using data from real experiments; advanced students may choose to use their own data. The students will develop understanding through constant presentation of their work and dialectical reflection over their choices, results, and interpretations.

This course is recommended for PhD and Master Students that are involved in neuroscience research projects. It is an advantage if the student has knowledge in algebra, linear algebra, statistics, calculus and neurophysiology. It is also recommended that the student is proficient in programming in Python.

After completing the course, the student should have the following overall learning outcomes defined in terms of knowledge, skills and general competence:

Knowledge

Upon successful completion of this course, the student knows:

• theoretical and practical aspects of data management and data processing for different data types (electrophysiology, imaging, and behavior), so they can independently engage with the state-of-the art literature.
• how statistics, dimensionality reduction, signal processing, calculus, and supervised and unsupervised learning may be applied to data analysis in neuroscience research.
• commonly encountered problems in data science and the most trusted strategies to solve them.
• basic data visualization techniques and how to employ them in exploratory data analysis and scientific communication.

Skills

Upon successful completion of this course, the student can:

• apply data analysis techniques to formulate a hypothesis, collect data, preprocess it, analyze it, and reach conclusions about the data.
• plan, design and implement data analysis pipelines for the most common types of data in neuroscience.
• identify and define a problem and craft a solution using data analytics.
• critically assess the results as well as justify and explain the methodological choice.
• identify new opportunities for organizational change including process improvements, cost reduction, or efficiency improvements.

General competence

Upon successful completion of this course the student can apply:

• data analysis principles to neuroscientific data in the context of its own research.
• methods and tools for data analysis and visualization.
• heuristics and strategies commonly used in the research field to solve data analysis problems.

The teaching methodology is oriented by Bloom's taxonomy of educational goals, namely, recollection, understanding, application, analysis, evaluation, and creativity. To promote recollection, understanding, and application, the course will consist of seminars taught by the teaching staff of OsloMet and other guests (experts in neuroscience or data science fields), coding workshops and problem solving oriented projects. Students will actively participate by implementing the full data processing pipeline from extracting the raw data to building visualizations. The pipeline and good habits will be consolidated through repetition in different modules, contexts, and data types, which is known to promote generalization of the knowledge. Organized in pairs, the students will constantly have the opportunity to recollect, explain the content to each other, and justify their work, as well as to provide feedback to their partners. With the intent to prepare the students to go beyond the methods taught in the course, once per module, the participants will read relevant papers in neuroscience, and discuss how to implement their analysis.

The students will have an assignment at the end of each module (4 in total) to consolidate the material learned in that week. All assignments need to be submitted in order to have access to the exam and to pass the course.

Final examination:

• An individual oral presentation, which counts for 40% of the final mark
• An individual project described in a final 3,000 to 5,000 word report.which counts for 60% of the final mark.

Each exam must be assessed to E or better for the course as a whole to be given a final grade.

The oral presentation cannot be appealed.

Only personal notes written in non-digital media will be allowed during the oral exam. All support materials are allowed in the other exams.

The final assessment will be graded on a grading scale from A to E (A is the highest grade and E the lowest) and F for fail.

Two examiners will be used, one of which can be external. External examiner is used regularly.

The course has 5 ECTS of overlapping content towards "Neuro-insights: Data Science Approaches in Neuroscience I" (STKD6800).