Authoring

ReactionNetwork

The static network container (ADR 0015: renamed from the ACSets-lineage ReactionNetworkSchema — it is a populated network INSTANCE, not the schema; the type-level object model is const SCHEMA). It is the inert, typed struct-of-columns store an authored model compiles to before it is handed to ReactionNetworkProblem for simulation. counts counts rows per object; columns is the NamedTuple of typed columns in ALLATTRS order; reactants is the promoted ReactantSpec incidence table (ADR 0003 Phase 2), populated lazily/on-merge (empty for a freshly-constructed or not-yet-promoted model — the runtime never reads it).

Macro that takes an expression corresponding to a reaction network and outputs a ReactionNetwork (the static struct-of-columns model), ready to pass to ReactionNetworkProblem for simulation.

Most arrows accepted (both right, left, and bi-drectional arrows). Use 0 or ∅ for annihilation/creation to/from nothing.

Custom functions and sampleable objects can be used as numeric parameters. Note that these have to be accessible from ReactiveDynamics's source code.

Examples

net = @reaction_network begin
    1.0, X ⟶ Y
    1.0, X ⟶ Y, priority => 6.0, prob => 0.7, capacity => 3.0
    1.0, ∅ --> (Poisson(0.3γ)X, Poisson(0.5)Y)
    (XY > 100) && (XY -= 1)
end
@push net 1.0 X ⟶ Y
@prob_init net X = 1 Y = 2 XY = α
@prob_params net γ = 1 α = 4

Add reactions to a network.

Examples

@push sir β * S * I * tdecay(@time()) S + I --> 2I name => SI2I
@push sir begin
    ν * I, I --> R, name => I2R
    γ, R --> S, name => R2S
end

Add new species to a model.

Examples

@add_species net S I R

Set species modality.

Supported modalities

• nonblock
• conserved
• rate

Examples

@mode net (r"proj\w+", r"experimental\w+") conserved
@mode net (S, I) conserved
@mode net S conserved

Alias an object name in a network.

Default names

Examples

@aka net species = resource transition = reaction

Set name of a transition in the model.

Examples

@name_transition net 1 = "name"
@name_transition net name = "transition_name"
@name_transition net "name" = "transition_name"

@append_transitions net transitions

Append a runtime-built collection of reaction lines to an existing net. transitions evaluates to a collection of strings, each one reaction line in the @reaction_network surface syntax; they are joined into a single begin…end block, parsed, and handed to @push. Use this when the set of transitions is assembled programmatically (a vector built in a loop, read from a table) rather than written literally — the literal-authoring path is @push.

Examples

lines = ["ν * I, I --> R, name => I2R", "γ, R --> S, name => R2S"]
@append_transitions net lines

Set cost.

Examples

@cost model experimental1=2 experimental2=3

Set reward.

Examples

@reward model experimental1=2 experimental2=3

Set valuation.

Examples

@valuation model experimental1=2 experimental2=3

Set initial values of species in a network.

Examples

@prob_init net X = 1 Y = 2 Z = h(α)
@prob_init net [1.0, 2.0, 3.0]

Set uncertainty in initial values of species in a network (stderr).

Examples

@prob_uncertainty net X = 0.1 Y = 0.2
@prob_uncertainty net [0.1, 0.2]

Set parameter values in a network.

Examples

@prob_params net α = 1.0 β = 2.0

Set model metadata (e.g. solver arguments)

Examples

@prob_meta net tspan = (0, 100.0) schedule = schedule_weighted!
@prob_meta sir tspan = 250 dt = 1   # `tstep` is a deprecated alias for `dt`

Add a periodic callback to a model.

Examples

@periodic net 1.0 X += 1

Add a jump process (with specified Poisson intensity per unit time step) to a model.

Examples

@jump net λ Z += rand(state.rng, Poisson(1.0))

Register a host function/definition into ReactiveDynamics scope so model expressions may call it by name (e.g. a custom rate/guard helper). Evaluates ex at macro-expansion time — an AUTHORING-time escape hatch, distinct from the eval-free runtime; do not use it to inject per-run data.

Examples

@register bool_cond(t) = (100 < t < 200) || (400 < t < 500)
@register tdecay(t) = exp(-t / 10^3)