Computational biology builds models of living systems, from the folding of one protein to the dynamics of a cell, a tissue or a population. Where bioinformatics handles biological data, this field asks how the system works: it writes equations and simulations, fits them to measurements, and predicts what would happen if something changed.
Key concepts
- Mechanistic models: differential equations for reactions, gene regulation and growth, with parameters taken from experiment.
- Stochastic models: inside a cell, molecule counts are small, so chance matters and simulations must sample it.
- Networks: metabolic, regulatory and interaction graphs, and what their structure implies.
- Structure and dynamics: how a molecule's shape and motion decide its function.
- Parameter fitting: estimating unknowns from noisy data, and asking which of them the data can really determine.
Vocabulary
- Systems biology: studying an organism as interacting parts rather than one gene at a time.
- Flux: the rate at which material flows through a reaction or pathway.
- Population dynamics: models of how numbers of individuals or species change over time.
- Identifiability: whether the data can pin down a parameter at all.
What to study, in order
- Introductory biology and enough calculus to read a differential equation.
- Simulate a growth model and a predator-prey model, and explore how parameters change the outcome.
- Write a stochastic simulation of a small reaction network and compare it with the deterministic version.
- Fit a model to a real or published dataset, and report the uncertainty of each parameter.
- Build and analyse a small regulatory network, and test which components control its behaviour.
Related tags
#bioinformatics #simulation #probability #optimization #computational-neuroscience