Tracking graph metrics for modeling the autonomic nervous system as a network
Sandya Subramanian, Professor
Computational Precision Health
Applications for Spring 2026 are closed for this project.
Our lab proposes to model the autonomic nervous system - the system in the body that "keeps the lights on" so to speak - as a connected and dynamic network. To do this, we need to design metrics to summarize the state of the network succinctly so that we can track changes in structure and function over time. We also need to setup computational infrastructure to compute these graphs from large continuous human datasets. For this, we are looking for a URAP intern to help design metrics to track graph dynamics over time and a second URAP intern to help setup a computational infrastructure to compute graphs from 72 hours of continuous data and assess these metrics.
Role: Tasks (focus on metrics):
1. Develop 3-5 key metrics that track the most salient characteristics of graph dynamics for the autonomic nervous system
2. Apply metrics to real-world data from human subjects to summarize graph dynamics and trends
3. Iterate on metric definition based on results
Learning outcomes
1. Apply concepts from mathematics, statistics, and graph theory to a real-world problem
2. Understand the intricacies and nuances of real-world human data
3. Bring together concepts from computation and biology together at an interdisciplinary interface
Tasks (focus on infrastructure):
1. Help PI setup and troubleshoot Savio workspace for P3 data from human subjects
2. Design pipelines to compute dynamic graphs from long continuous stretches of data from human subjects, accounting for user-defined hyperparameters for different use cases and handling missing data etc.
3. Test pipelines on existing datasets in the lab and iterate
Learning outcomes:
1. Understand the challenges of handling real-world big data, especially with privacy and confidentiality constraints
2. Understand and apply software engineering best practices to allow pipelines to be both robust and flexible for maximal use across users and contexts
Qualifications: Metrics-focused position
1. Have taken mathematics and statistics courses including linear algebra, differential equations, upper level probability and statistics
2. Have a working understanding of graph theory, either from coursework or other experiences
3. Have taken courses in or worked with signal processing for time series
4. 2+ years coding experience in Python
5. Comfortable with mathematical theory and derivations, including working out your own
6. Experience collaborating in a team
7. Solid time management skills, even during weeks with midterms/other assignments
Infrastructure-focused position
1. Have taken foundational computer science coursework, including CS61 series
2. Have taken mathematics and statistics courses including linear algebra, differential equations, upper level probability and statistics
3. Have taken courses in or worked with signal processing for time series
4. 3+ years coding experience in Python, including developing pipelines for use by others
5. Experience collaborating in a team
6. Solid time management skills, even during weeks with midterms/other assignments
Hours: 9-11 hrs
Related website: https://www.subramanianlab.com
Mathematical and Physical Sciences Engineering, Design & Technologies Biological & Health Sciences