Skip to main content

Crate tpt_sci_reaction_network

Crate tpt_sci_reaction_network 

Source
Expand description

§tpt-sci-reaction-network

A species / rate / stoichiometry DSL for compartmental models — the Rust equivalent of Julia’s Catalyst.jl — built from scratch for the tpt-science pillar.

Given a chemical reaction network (species, reactions, and rate laws), the crate compiles the law of mass action into the deterministic ODE

dy/dt = S · r(y, p)

where S is the stoichiometry matrix and r the per-reaction rate vector, and hands the result to tpt_sci_ode for integration. It also exposes the stoichiometry matrix and rate vector directly for inspection, parameter estimation, or coupling into a larger multi-scale model.

In addition to the deterministic mass-action ODE, the crate provides three stochastic backends operating on the same IR: exact trajectories via ReactionSystem::simulate_ssa (Gillespie’s direct method), approximate but much faster trajectories via ReactionSystem::simulate_tau_leaping (explicit tau-leaping, Poisson-distributed reaction counts per step), and the diffusion-limit ReactionSystem::simulate_cle (the Chemical Langevin Equation, Euler–Maruyama with Gaussian noise) — see crate::tau_leap for the tradeoffs between them.

Beyond simulation, ReactionSystem::conservation_laws finds linear conservation laws (the left null space of the stoichiometry matrix) for structural analysis of a network, and ReactionNetwork::from_sbml reads a practical subset of SBML XML directly into this crate’s IR (see crate::sbml for exactly what subset).

This is the v1 scope from spec.txt / registry.toml: a CRN IR, a mass-action ODE builder, exact/tau-leaping/CLE stochastic backends, a minimal SBML reader, and conservation-law detection. (The previous plan deferred the stochastic backend to wrapping rebop; it is now implemented from scratch instead.)

§Building models

Two equivalent ways to define a model:

  1. Programmatically with ReactionNetwork, registering species and parameters by name and appending reactions with explicit indices.
  2. Textually with the Catalyst.jl-style ReactionNetwork::from_dsl, e.g. kB, S + E --> SE.

§Example

Michaelis–Menten enzyme kinetics (S + E ⇌ SE → P + E):

use tpt_sci_ode::Method;
use tpt_sci_reaction_network::ReactionNetwork;

let model = ReactionNetwork::from_dsl(
    "kB, S + E --> SE
     kD, SE --> S + E
     kP, SE --> P + E",
).unwrap();
let mut model = model;
model.set_parameter("kB", 0.01).unwrap();
model.set_parameter("kD", 0.1).unwrap();
model.set_parameter("kP", 0.1).unwrap();

// E + SE is conserved: total enzyme stays at its initial 10.
let y0 = model.initial_state(&[("S", 50.0), ("E", 10.0)]).unwrap();
let prob = model.to_ode_problem(&y0, 0.0).unwrap();
let y = prob.solve(Method::Bdf, 200.0).unwrap();

let e = model.species_index("E").unwrap();
let se = model.species_index("SE").unwrap();
assert!((y[e] + y[se] - 10.0).abs() < 1e-6);

Licensed under either of MIT or Apache-2.0 at your option.

Re-exports§

pub use error::ReactionNetworkError;
pub use model::RateLaw;
pub use model::Reaction;
pub use model::ReactionNetwork;
pub use model::ReactionSystem;
pub use model::Term;
pub use sbml::SbmlModel;
pub use ssa::SsaTrajectory;
pub use tau_leap::TauLeapConfig;

Modules§

analysis
Stoichiometric network analysis: linear conservation-law detection.
error
model
ode
sbml
Minimal SBML (Systems Biology Markup Language) reader.
ssa
Stochastic simulation (Gillespie’s direct method) for reaction networks.
tau_leap
Explicit tau-leaping and the Chemical Langevin Equation (CLE): two approximate stochastic integrators that complement the exact SSA in crate::ssa.