Quantifying the Flow of Information
Preface
This website contains the full text of my PhD Thesis Quantifying the Flow of Information converted and adapted from the original LaTeX source. In this form it is much easier to read, especially on mobile.
Use the sidebar on the left to navigate the chapters.
About
How do living cells information? Which parts of signals carry information? This thesis develops Path Weight Sampling (PWS), a novel Monte Carlo technique that—for the first time—makes it possible to compute the mutual information between input and output trajectories exactly for any stochastic system.
The key idea is to use the master equation or the Onsager-Machlup functional to evaluate the exact conditional probability of an individual output trajectory for a given input trajectory, and to average this via Monte Carlo sampling in trajectory space. By establishing connections between information theory and statistical physics (such as path probabilities and partition functions), we leverage techniques from soft condensed-matter physics to make this computation efficient.
In the thesis, we also demonstrate the practical utility PWS by applying it to bacterial chemotaxis, one of the best-characterized signaling systems in biology.
Software
The methods developed in this thesis are implemented in a Julia package: PathWeightSampling.jl
Resources
- PDF version (9.5 MB)
- GitHub repository
The research was conducted between 2019 and 2025 at AMOLF, with funding from the Dutch Research Council (NWO).
Acknowledgments
The research was conducted between 2019 and 2025 in the Biochemical Networks group of Pieter Rein ten Wolde at AMOLF, with funding from the Dutch Research Council (NWO) and the European Research Council (ERC).
License
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
Citation
If you use this thesis or parts of it in your own work, please cite it as:
Reinhardt, M. J. (2025). Quantifying the Flow of Information. PhD Thesis, Vrije Universiteit Amsterdam. https://doi.org/10.5463/thesis.1092.
Contact
For any questions or feedback regarding this thesis, please contact me at mail@manuel-rhdt.de.