We are pleased to announce the Physalia online course: Machine Learning Methods for Longitudinal Data with Python.

Dates: 6th-9th May

Course website: https://www.physalia-courses.org/courses-workshops/longitudinal-data/

This course will introduce machine learning methods for analyzing longitudinal (sequence) data, where time and cause-effect relationships are important. You will learn how to handle the specific challenges of working with this type of data, from visualization and modeling to interpretation.

Course Highlights:

• Understand how time and causation affect data analysis

• Learn to identify and address biases such as confounding and mediator effects

• Apply machine learning methods to sequence data

• Use graph models, Bayesian networks, and time-series forecasting

• Work with real-world biological datasets, including epidemiology and gene expression

Who Should Attend?

This course is designed for advanced students, researchers, and professionals working with biological data that changes over time. A basic understanding of Python and Linux is helpful but not required.

Course Format:

The course is structured over four days and includes lectures, discussions, and hands-on practical exercises using Python, Jupyter Notebooks, and the Linux command line. Participants will work on exercises, interpret results, and discuss their own research questions.

Schedule (Berlin time):

Day 1 (2-8 PM): Introduction to sequence data, statistical models, bias handling

Day 2 (2-8 PM): Graph models, Bayesian networks, ML approaches for time-series prediction

Day 3 (2-8 PM): Longitudinal data in epidemiology, deep learning, Transformer models

Day 4 (2-8 PM): Model diagnostics, multi-omics case study, final discussion

For the full list of our courses and workshops, please visit: https://www.physalia-courses.org/courses-workshops/