From calcium imaging to graph topology.

Area of Science:

• Computational neuroscience
• Systems neuroscience
• Data analysis

Background:

• Advances in microscopy and protein engineering enable whole-brain neural activity imaging in behaving animals.
• Exponential growth in simultaneously recorded neurons presents significant data analysis challenges.
• Existing computational methods often use disparate terminology and mathematical concepts.

Purpose of the Study:

• To collect, organize, and explain diverse data analysis techniques for whole-brain imaging.
• To demonstrate these methods using larval zebrafish as a model system.
• To highlight the applicability of these computational approaches to broader biological problems.

Main Methods:

• Linear regression for analyzing relationships between two variables.
• Network science and applied topology for understanding patterns among multiple variables.
• Generative models for generating hypotheses on neural network structure and dynamics.

Main Results:

• Demonstration of various computational methods applied to whole-brain imaging data.
• Illustration of how network science and topology can reveal complex neural relationships.
• Exploration of generative models for predicting neural wiring and disease progression.

Conclusions:

• The presented computational approaches are suitable for population-scale neural network modeling.
• These methods extend beyond larval zebrafish, applicable to various systems neuroscience research.
• Computational techniques from network science and topology offer broad applications across biological sciences.