DataSAI for Neuroscience Summer School
About DataSAI for Neuroscience Summer School 2025
Welcome to DataSAI, an immersive two-week summer school to train neuroscientists in data science, statistical inference, and machine learning. The summer school runs June 16–27, 2025. This webpage has all of the materials for the summer school.
Personnel
- Course instructor: Justin Bois is a teaching professor in the Division of Biology and Biological Engineering at Caltech, where he teaches a variety of courses, including courses on data analysis in the biological sciences.
- Teaching assistant: Kevin Le is a Ph.D. student in the Computation and Neural Systems option at Caltech. He works in the labs of Ueli Rutishauser and Pietro Perona where he studies how meaning is represented in neural activity.
- Teaching assistant: Adi Nair is a postdoc at Caltech working with David Anderson, Pietro Perona, and Scott Linderman (at Stanford). He uses dynamical systems modeling or neural circuits to understand emotion. Adi conceived and organized the first DataSAI summer school in 2022.
Course structure and schedule
We meet every day, 9–5, save for June 19, in honor of the Juneteenth holiday, for a total of nine days. Each day has a morning session and an afternoon session, with each session dedicated to a given topic. Each half-day session is further split into an instructional and practical section. In the instructional section, topics are introduced and discussion in a lecture and/or follow-along format. In the practical sections, students apply the concepts in exercises. The topics of the sessions are below.
- Data practicalities
- Mon, June 16 AM: Polars and split-apply-combine
- Mon, June 16, PM: Data display
- Tue, June 17, AM: Data file formats
- Theory
- Tue, June 17, PM: Probability review
- Wed, June 18, AM: Sampling out of probability distributions
- Wed, June 18, PM: Sampling with Markov chain Monte Carlo
- Fri, June 20, AM: Bayesian modeling and inference (with prior predictive checks)
- Techniques
- Fri, June 20, PM: Statistical inference by Markov chain Monte Carlo
- Mon, June 23, AM: Model assessment and principled workflows
- Mon, June 23, PM: Summarizing posteriors with optimization
- Tue, June 24, AM: The expectation-maximization algorithm and mixture models
- Specific models
- Tue, June 24, PM: Hierarchical models
- Wed, June 25, AM: Principal component analysis
- Wed, June 25, PM: Probabilistic PCA and factor analysis
- Thu, June 26, AM: Hidden Markov models
- Thu, June 26, PM: Generalized linear models
- Fri, June 27, AM: Variate-covariate models
- Fri, June 27, PM: Data smoothing and Gaussian processes
Copyright and License
Copyright 2025, Justin Bois.
With the exception of pasted graphics, where the source is noted, this work is licensed under a Creative Commons Attribution License CC BY-NC-SA 4.0. All code contained herein is licensed under an MIT license.