Course onSystems Biology & Mathematical Modeling

• What the parts are. Protein–protein interaction, signalling, gene regulatory and metabolic networks — what a node and an edge actually mean in each one.
• Where the data comes from. The major databases, what kind of evidence sits behind each entry, and how much weight it will carry.
• Reading structure. Degree, centrality, paths and modules — finding the important nodes without knowing anything about dynamics yet.
• Recurring patterns. Feedback loops and feed-forward loops, and the behaviour each one produces.
• Building your own. Top-down inference from expression data, and bottom-up assembly from the primary literature.

You produce: two reconstructed networks of your own — one built by hand from the literature, one inferred from an RNA-seq count matrix — with their structural analysis.

• Writing down a rate. Mass action, Michaelis–Menten and the Hill function — what each one assumes about the biology.
• Assembling the system. Turning a set of connected reactions into a system of differential equations.
• Getting the numbers. Where parameters come from — dose–response curves, time courses, binding measurements — and how to fit them when nothing is published.
• Solving it. Analytically, at steady state, and numerically over time.
• Checking it. Simulating knockouts and perturbations, then testing the model against data it was never fitted to.

You produce: a parameterised ODE model of a regulatory circuit, simulated and tested against independent data.

• Why metabolism needs its own approach. Thousands of reactions, almost no measured kinetics — and what to do about it.
• The assumptions. Steady state, mass balance and limits on individual reactions: what each one buys you, and what it costs you.
• The optimisation. Choosing what the cell is assumed to be maximising, and how that single choice shapes every answer you get.
• The analyses. Predicting growth, comparing nutrients and media, screening gene and enzyme knockouts, switching between aerobic and anaerobic conditions.
• Reading the result. What a predicted flux distribution supports as a claim, and where its limits are.

You produce: an analysis of a whole-organism metabolic model — a knockout screen, a media comparison, and an interpretation you can defend.