| title |
|---|
Syllabus for 'Data science in Biomedical research: Reproducible quantitative analytics and pipelines' |
Biomedical data includes a diverse range of data types from patient data from national registeries, to biomarker data, to genetic and sequence data, to electrophysiological recordings from human or animal models, to simulations, often covering multiple spatio-temporal scales. These data are usually high dimensional, noisy, and fragmented. Handling such data is challenging and understanding it is an important first step in a data-driven approach to build predictive or inferential models. This course introduces techniques to analyze biomedical data using the Python programing language. The focus of the course is to integrate reproducible and open scientific research principles such as the use of code to generate the results, having well-documented modular coding, correctness of procedure and chronology of execution, being transparent and open throughout the entire research process, and correct and cautious interpretation and presentation of the results.
- To develop a proficiency in coding and doing data analysis in Python.
- To recognize the importance of and to apply "tidy data" principles.
- To write reproducible, well-documented, and modular Python code.
- To learn how to munge, wrangle, and management data reproducibly in Python.
- To apply common statistical techniques to biological data, particularly high dimensional data.
- To recognize the importance of and to ensure reproducibility in documents such as manuscripts and theses.
- To generate publication quality outputs such as figures and documents that effectively communicate technical content.
- To work in a productive and collaborative environment in a team-based project.
Graduate students that are from different backgrounds and have entered either neuromodulation, neuroscience, or biomedical engineering programs.
- No programming experience in necessary.
Multiple, small assignments will be given out to test and reinforce learning of the material. A final team-based project will assess and reinforce students' grasp of the material that was taught throughout the course.
The final project will be to obtain an open dataset, formulate a research hypothesis, analyze the data, and write up a document, all while adhering to reproducible and open scientific guidelines. The project will include creating a project plan that includes scope, rationale, deliverables, and milestones. For a project to be successfully completed, we expect that:
- Team members have discussed and decided as a team on the dataset and on the roles of the members.
- Set up and (regularly) use a project repository on GitHub for the code and document.
- Python code is properly documented, structured, and written.
- Data has been properly wrangled and converted into a tidy format.
- The report is written in a Jupyter Notebook and that the output and results are completely reproducible.
The progress of the project will be checked and assistance will be provided where needed.