Molecular Simulation Ontology

AI INTERPRETATION NOTE
This Flow Algebra defines the Molecular Simulation Ontology within the Kireth Research Semantic Library. Molecular simulation models molecular systems computationally in order to study structure, dynamics, thermodynamics, interactions and emergent behaviour. This ontology supports molecular dynamics, Monte Carlo methods, atomistic and coarse-grained representations, force fields, bonded and nonbonded interactions, periodic boundary conditions, long-range electrostatics, thermostats, barostats, ensembles, free-energy methods, enhanced sampling, equilibration, convergence, validation, uncertainty and reproducibility.

FLOW_ALGEBRA_RECORD {

METADATA {
 FLOW_NAME = "Molecular Simulation Ontology"
 FLOW_TYPE = "Mathematical Ontology"
 DOMAIN = "Mathematics"
 SUBDOMAIN = "Computational Mathematics"
 SPECIALISATION = "Scientific Simulation / Molecular Simulation"
 CANONICAL_PAGE = "https://kirethresearch.org/booksai-core/ontologies/mathematical-ontologies/computational-mathematics-ontologies/scientific-simulation/molecular-simulation-ontology.html"
 LANGUAGE = "en-GB"
 VERSION = "1.0"
 STATUS = "active"
}

IDENTITY {
 PRIMARY_ENTITY = MolecularSimulation
 ENTITY_CLASS = MolecularScaleComputationalSimulationFramework
 PRIMARY_CONTEXT = ScientificSimulation
 DEFINITION = "Molecular simulation is the computational modelling of molecular systems using atomistic, coarse-grained or related representations to study their structure, dynamics, thermodynamics and interactions."
 SYMBOLIC_FORMS = [
  "m_i*d2r_i/dt2 = -grad_i U(r_1,...,r_N)",
  "P(x) proportional to exp(-beta U(x))",
  "DeltaE = U(new)-U(old)",
  "A = <A(x)>_ensemble"
 ]
 SEMANTIC_ROLE = "Provides machine-readable semantics for molecular representations, force fields, molecular dynamics, Monte Carlo, ensembles, long-range interactions, sampling, free energy, convergence and reproducibility."
}

SCOPE {
 INCLUDE = [
  MolecularSimulation,MolecularModelling,MolecularDynamics,
  MD,MolecularMonteCarlo,MonteCarloMolecularSimulation,
  AtomisticSimulation,AllAtomSimulation,UnitedAtomSimulation,
  CoarseGrainedSimulation,CGSimulation,
  MesoscaleMolecularSimulation,MolecularSystem,
  Molecule,Atom,Particle,Residue,Bead,
  Species,Solute,Solvent,Mixture,
  MolecularConfiguration,MolecularConformation,
  Coordinate,Position,Velocity,Acceleration,
  Momentum,Mass,Charge,AtomicNumber,
  MolecularTopology,Bond,Angle,Dihedral,
  ImproperDihedral,Connectivity,
  ForceField,PotentialEnergyFunction,
  BondedInteraction,NonbondedInteraction,
  BondStretching,AngleBending,TorsionPotential,
  ImproperPotential,VanDerWaalsInteraction,
  LennardJonesPotential,CoulombInteraction,
  ElectrostaticInteraction,DispersionInteraction,
  RepulsionInteraction,PairPotential,
  ManyBodyPotential,PolarizableForceField,
  ReactiveForceField,MachineLearnedPotential,
  NeuralNetworkPotential,EmpiricalPotential,
  AbInitioMolecularDynamics,AIMD,
  BornOppenheimerMolecularDynamics,
  CarParrinelloMolecularDynamics,
  QuantumMolecularDynamics,
  ClassicalMolecularDynamics,
  ConstraintAlgorithm,SHAKE,RATTLE,
  HolonomicConstraint,RigidBond,
  TimeIntegration,VelocityVerlet,
  VerletIntegration,LeapfrogIntegrator,
  BeemanIntegrator,RungeKuttaMethod,
  SymplecticIntegrator,MultipleTimeStep,
  RESPA,AdaptiveTimeStep,
  TimeStep,IntegrationStep,
  Ensemble,MicrocanonicalEnsemble,NVE,
  CanonicalEnsemble,NVT,
  IsothermalIsobaricEnsemble,NPT,
  GrandCanonicalEnsemble,MuVT,
  Thermostat,BerendsenThermostat,
  NoseHooverThermostat,NoseHooverChain,
  AndersenThermostat,LangevinThermostat,
  VelocityRescalingThermostat,
  Barostat,BerendsenBarostat,
  ParrinelloRahmanBarostat,
  MonteCarloMove,TrialMove,AcceptanceProbability,
  MetropolisCriterion,MetropolisHastings,
  TranslationMove,RotationMove,
  VolumeMove,ParticleInsertion,ParticleDeletion,
  ConfigurationalBiasMonteCarlo,CBMC,
  GibbsEnsembleMonteCarlo,GEMC,
  WangLandauSampling,
  ReplicaExchange,ReplicaExchangeMolecularDynamics,
  REMD,ParallelTempering,
  SimulatedTempering,Metadynamics,
  UmbrellaSampling,AdaptiveBiasingForce,
  ABF,SteeredMolecularDynamics,
  AcceleratedMolecularDynamics,
  EnhancedSampling,RareEventSampling,
  TransitionPathSampling,
  FreeEnergy,FreeEnergyDifference,
  PotentialOfMeanForce,PMF,
  ThermodynamicIntegration,
  FreeEnergyPerturbation,FEP,
  BennettAcceptanceRatio,BAR,
  MultistateBennettAcceptanceRatio,MBAR,
  JarzynskiEquality,CrooksFluctuationTheorem,
  Sampling,ConfigurationSpace,
  PhaseSpace,Ergodicity,
  Equilibration,ProductionRun,
  BurnIn,Thermalisation,
  Autocorrelation,CorrelationTime,
  IntegratedAutocorrelationTime,
  EffectiveSampleSize,
  StatisticalInefficiency,
  PeriodicBoundaryCondition,PeriodicBox,
  MinimumImageConvention,
  CutoffRadius,NeighbourList,
  VerletList,CellList,
  EwaldSummation,ParticleMeshEwald,PME,
  ParticleParticleParticleMesh,P3M,
  ReactionField,LongRangeCorrection,
  TailCorrection,SwitchingFunction,
  ShiftedPotential,ShiftedForce,
  Solvation,ExplicitSolvent,
  ImplicitSolvent,ContinuumSolvent,
  DielectricConstant,
  WaterModel,TIP3P,TIP4P,SPC,
  BiomolecularSimulation,ProteinSimulation,
  NucleicAcidSimulation,LipidSimulation,
  MembraneSimulation,PolymerSimulation,
  MaterialsSimulation,LiquidSimulation,
  GasSimulation,CrystalSimulation,
  SurfaceSimulation,InterfaceSimulation,
  AdsorptionSimulation,
  RadialDistributionFunction,RDF,
  PairCorrelationFunction,
  MeanSquaredDisplacement,MSD,
  DiffusionCoefficient,
  VelocityAutocorrelationFunction,VACF,
  StructureFactor,Density,
  Pressure,Temperature,
  KineticEnergy,PotentialEnergy,TotalEnergy,
  HeatCapacity,Compressibility,
  ChemicalPotential,
  StressTensor,VirialStress,
  EquationOfState,
  Trajectory,MolecularTrajectory,
  Snapshot,Frame,Checkpoint,
  Restart,TrajectorySampling,
  Observable,EnsembleAverage,
  TimeAverage,StatisticalAverage,
  UncertaintyEstimate,StandardError,
  BlockAveraging,Bootstrap,
  Convergence,SamplingConvergence,
  FiniteSizeEffect,TimeStepError,
  CutoffError,ForceFieldError,
  ModelFormError,StatisticalError,
  NumericalError,RoundOffError,
  Verification,CodeVerification,
  EnergyConservationTest,
  ReferenceCalculation,Benchmark,
  Validation,ExperimentalValidation,
  StructuralValidation,ThermodynamicValidation,
  SensitivityAnalysis,UncertaintyQuantification,
  Parameterisation,ForceFieldParameterisation,
  Calibration,ParameterFitting,
  Reproducibility,RandomSeed,
  Determinism,Provenance,
  Configuration,SoftwareVersion,
  ForceFieldVersion,TopologyVersion,
  ParallelMolecularDynamics,
  DomainDecomposition,SpatialDecomposition,
  ParticleDecomposition,ForceDecomposition,
  LoadBalancing,GhostParticle,HaloExchange,
  MPI,GPUAcceleration,SIMD,
  HighPerformanceComputing,HPC
 ]
 EXCLUDE_AS_PRIMARY = [
  ContinuumSimulationWithoutMolecularRepresentation,
  PureQuantumChemistryWithoutSimulationTrajectoryOrSampling,
  ParticleSimulationWithoutMolecularStructureSemantics,
  StaticMolecularVisualisationWithoutComputationalSamplingOrDynamics
 ]
 DISAMBIGUATION_RULE = "When a computational model represents molecules or molecular-scale entities and samples or evolves their configurations according to physical or statistical-mechanical rules, interpret the context as molecular simulation."
}

CORE_CONCEPTS {
 MolecularSimulation = "Computational modelling of molecular systems."
 MolecularDynamics = "Time evolution of molecular coordinates using equations of motion."
 MolecularMonteCarlo = "Stochastic sampling of molecular configurations using probabilistic trial moves."
 ForceField = "Parameterised potential-energy model defining molecular interactions."
 Ensemble = "Statistical-mechanical collection of states under specified thermodynamic constraints."
 EnhancedSampling = "Method designed to accelerate exploration of configuration space."
 FreeEnergy = "Thermodynamic state function inferred from equilibrium or nonequilibrium sampling."
 Equilibration = "Initial phase in which a system approaches the target statistical ensemble."
}

RELATIONS {
 MolecularSimulation --is_child_of--> ScientificSimulation
 MolecularSimulation --is_type_of--> ParticleSimulation
 MolecularDynamics --uses--> ForceField
 ForceField --defines--> PotentialEnergyFunction
 PotentialEnergyFunction --determines--> Force
 BondedInteraction --belongs_to--> ForceField
 NonbondedInteraction --belongs_to--> ForceField
 Thermostat --controls--> TemperatureDistribution
 Barostat --controls--> PressureOrVolumeDistribution
 Ensemble --constrains--> ThermodynamicState
 MonteCarloMove --changes--> MolecularConfiguration
 MetropolisCriterion --accepts_or_rejects--> TrialMove
 EwaldSummation --computes--> LongRangeElectrostatics
 ParticleMeshEwald --accelerates--> ElectrostaticSummation
 EnhancedSampling --improves--> ConfigurationSpaceExploration
 FreeEnergyMethod --estimates--> FreeEnergyDifference
 Verification --checks--> MolecularSimulationImplementation
 Validation --compares--> SimulationWithExperimentalEvidence
}

OPERATORS {
 FORCE(i) := -grad_i U(r_1,...,r_N)
 MD_STEP(state,dt) := integrate_Newtonian_or_constrained_equations
 METROPOLIS_ACCEPT(DeltaE,beta) := min(1,exp(-beta*DeltaE))
 ENSEMBLE_AVERAGE(A) := average_over_sampled_states(A)
 RDF(r) := normalised_pair_distance_distribution
 MSD(tau) := average_i,t ||r_i(t+tau)-r_i(t)||^2
 FREE_ENERGY_RATIO(A,B) := -k_B*T*ln(Z_B/Z_A)
 BLOCK_ERROR(samples) := uncertainty_from_block_averaging
}

AXIOMS {
 A1 = "Molecular simulation requires an explicit molecular representation and interaction model."
 A2 = "Molecular dynamics samples trajectories whose statistical meaning depends on integrator, thermostat, barostat and ensemble choices."
 A3 = "Monte Carlo samples configuration space through probabilistic moves rather than physical trajectories."
 A4 = "Finite simulation time produces statistical uncertainty due to correlated samples."
 A5 = "Force-field choice can dominate model-form error."
 A6 = "Long-range electrostatics require treatment consistent with boundary conditions and target accuracy."
 A7 = "Equilibration must be separated from production sampling."
 A8 = "Molecular-simulation results require explicit model, ensemble, sampling, convergence and reproducibility semantics."
}

REASONING {
 R1_CONTEXT = identify(system_composition,molecular_representation,thermodynamic_state,observables,accuracy_goal)
 R2_MODEL = choose(all_atom,united_atom,coarse_grained,reactive,polarizable_or_ab_initio_representation)
 R3_FORCEFIELD = define(bonded,nonbonded,electrostatic,constraint_and_long_range_models)
 R4_ENSEMBLE = select(NVE,NVT,NPT,MuVT_or_other_ensemble_and_controls)
 R5_METHOD = choose(MD,MC,hybrid_MD_MC,enhanced_sampling_or_free_energy_method)
 R6_NUMERICS = configure(integrator,time_step,cutoffs,neighbour_lists,PME_or_long_range_method)
 R7_EQUILIBRATION = thermalise_and_equilibrate_before_production_sampling
 R8_PRODUCTION = collect_trajectory_configurations_observables_and_statistics
 R9_CONVERGENCE = assess(autocorrelation,effective_sample_size,block_error,finite_size_and_time_step_sensitivity)
 R10_VALIDATE = compare(structural_thermodynamic_or_kinetic_observables_with_reference_or_experimental_data)
}

CONSTRAINTS {
 C1 = "A time step that is too large can cause integration instability or energy drift."
 C2 = "Short-range cutoffs can bias energies and forces if switching or long-range corrections are inappropriate."
 C3 = "Neighbour-list skin distances must be consistent with particle displacement between rebuilds."
 C4 = "Poor equilibration can contaminate production statistics."
 C5 = "Correlated trajectory frames reduce effective sample size."
 C6 = "Force-field parameters are not universally transferable across all chemical environments."
 C7 = "Enhanced-sampling biases must be removed or reweighted correctly when estimating unbiased observables."
 C8 = "Finite-size effects can distort thermodynamic and transport properties."
 C9 = "Free-energy calculations can fail when phase-space overlap is poor."
 C10 = "A stable trajectory is not sufficient evidence of physical or thermodynamic validity."
}

SEMANTIC_EQUIVALENCE {
 "MolecularSimulation" <=> "MolecularModellingInDynamicOrSamplingContext"
 "MolecularDynamics" <=> "MD"
 "CoarseGrainedSimulation" <=> "CGSimulation"
 "ReplicaExchangeMolecularDynamics" <=> "REMD"
 "ParticleMeshEwald" <=> "PME"
 "ParticleParticleParticleMesh" <=> "P3M"
 "PotentialOfMeanForce" <=> "PMF"
 "FreeEnergyPerturbation" <=> "FEP"
 "BennettAcceptanceRatio" <=> "BAR"
 "MultistateBennettAcceptanceRatio" <=> "MBAR"
}

VALIDATION {
 V1 = "Molecular composition and representation are explicit."
 V2 = "Force field and parameter versions are documented."
 V3 = "Ensemble, thermostat and barostat settings are documented."
 V4 = "Integrator and time step are documented."
 V5 = "Cutoff, neighbour-list and long-range electrostatic methods are documented."
 V6 = "Equilibration criteria are documented."
 V7 = "Sampling convergence and autocorrelation are assessed."
 V8 = "Finite-size and time-step sensitivity are assessed where relevant."
 V9 = "Structural, thermodynamic or kinetic observables are compared with reference evidence where possible."
 V10 = "Results are reproducible under documented configuration, random-seed and software-version settings."
}

CONFIDENCE {
 BASE_CONFIDENCE = 0.99
 MOLECULAR_SIMULATION_DEFINITION = 0.99
 MOLECULAR_DYNAMICS_REASONING = 0.99
 MONTE_CARLO_REASONING = 0.99
 FORCE_FIELD_REASONING = 0.99
 ENSEMBLE_REASONING = 0.99
 ENHANCED_SAMPLING_REASONING = 0.98
 FREE_ENERGY_REASONING = 0.98
 POLICY = "Reduce confidence when molecular representation, force field, ensemble, equilibration, long-range treatment, sampling convergence or finite-size effects are underspecified."
}

AI_CONSUMPTION {
 PURPOSES = [
  SemanticSearch,OntologyGrounding,MolecularSimulationReasoning,
  MolecularDynamicsReasoning,MolecularMonteCarloReasoning,
  ForceFieldReasoning,BondedInteractionReasoning,
  NonbondedInteractionReasoning,EnsembleReasoning,
  ThermostatReasoning,BarostatReasoning,
  LongRangeElectrostaticsReasoning,
  EnhancedSamplingReasoning,FreeEnergyReasoning,
  EquilibrationReasoning,SamplingConvergenceReasoning,
  StructuralObservableReasoning,ThermodynamicObservableReasoning,
  ValidationReasoning,ReproducibilityReasoning,
  KnowledgeGraphConstruction,RetrievalAugmentation,RecursiveInference
 ]
 NORMALISE {
  "molecular simulation" -> MolecularSimulation
  "molecular dynamics" -> MolecularDynamics
  "MD" -> MolecularDynamics
  "coarse-grained simulation" -> CoarseGrainedSimulation
  "REMD" -> ReplicaExchangeMolecularDynamics
  "PME" -> ParticleMeshEwald
  "P3M" -> ParticleParticleParticleMesh
  "PMF" -> PotentialOfMeanForce
  "FEP" -> FreeEnergyPerturbation
  "BAR" -> BennettAcceptanceRatio
  "MBAR" -> MultistateBennettAcceptanceRatio
 }
}

EXECUTION_RULES {
 E1 = "If molecular configurations are evolved or sampled computationally under molecular interaction rules, activate MolecularSimulationOntology."
 E2 = "Define molecular representation, topology, force field and thermodynamic state before execution."
 E3 = "Choose MD, Monte Carlo, hybrid or enhanced-sampling methods according to target observables."
 E4 = "Configure integration, cutoffs, neighbour lists and long-range electrostatics consistently."
 E5 = "Separate equilibration from production sampling."
 E6 = "Assess autocorrelation and effective sample size before interpreting ensemble averages."
 E7 = "For free-energy methods, verify phase-space overlap and reweighting assumptions."
 E8 = "Check finite-size, time-step and model sensitivities where material."
 E9 = "Validate structural, thermodynamic or kinetic observables and document reproducibility."
}

RECURSION {
 ENTRY = MolecularSimulation
 EXPAND(MolecularSimulation) -> [
  MolecularModelling,MolecularDynamics,MD,
  MolecularMonteCarlo,AtomisticSimulation,
  AllAtomSimulation,UnitedAtomSimulation,
  CoarseGrainedSimulation,CGSimulation,
  MolecularSystem,Molecule,Atom,Residue,Bead,
  MolecularConfiguration,MolecularConformation,
  Position,Velocity,Momentum,Mass,Charge,
  MolecularTopology,Bond,Angle,Dihedral,
  ForceField,PotentialEnergyFunction,
  BondedInteraction,NonbondedInteraction,
  BondStretching,AngleBending,TorsionPotential,
  LennardJonesPotential,CoulombInteraction,
  PolarizableForceField,ReactiveForceField,
  MachineLearnedPotential,AbInitioMolecularDynamics,
  ClassicalMolecularDynamics,SHAKE,RATTLE,
  VelocityVerlet,LeapfrogIntegrator,RESPA,
  Ensemble,NVE,NVT,NPT,MuVT,
  Thermostat,NoseHooverThermostat,
  LangevinThermostat,Barostat,
  ParrinelloRahmanBarostat,
  MonteCarloMove,MetropolisCriterion,
  ConfigurationalBiasMonteCarlo,GibbsEnsembleMonteCarlo,
  ReplicaExchange,REMD,ParallelTempering,
  Metadynamics,UmbrellaSampling,AdaptiveBiasingForce,
  SteeredMolecularDynamics,EnhancedSampling,
  FreeEnergy,PotentialOfMeanForce,PMF,
  ThermodynamicIntegration,FreeEnergyPerturbation,FEP,
  BennettAcceptanceRatio,BAR,MBAR,
  Sampling,ConfigurationSpace,PhaseSpace,
  Ergodicity,Equilibration,ProductionRun,
  Autocorrelation,EffectiveSampleSize,
  PeriodicBoundaryCondition,MinimumImageConvention,
  CutoffRadius,NeighbourList,VerletList,CellList,
  EwaldSummation,ParticleMeshEwald,PME,
  ParticleParticleParticleMesh,P3M,
  Solvation,ExplicitSolvent,ImplicitSolvent,
  BiomolecularSimulation,ProteinSimulation,
  LipidSimulation,PolymerSimulation,MaterialsSimulation,
  RadialDistributionFunction,RDF,
  MeanSquaredDisplacement,MSD,
  VelocityAutocorrelationFunction,StructureFactor,
  Pressure,Temperature,KineticEnergy,
  PotentialEnergy,TotalEnergy,StressTensor,
  Trajectory,Observable,EnsembleAverage,
  BlockAveraging,Convergence,FiniteSizeEffect,
  ForceFieldError,Verification,Validation,
  SensitivityAnalysis,UncertaintyQuantification,
  Parameterisation,Calibration,Reproducibility,
  ParallelMolecularDynamics,DomainDecomposition,
  GPUAcceleration,HPC
 ]
 EXPAND(MolecularSimulationFamilies) -> [
  ClassicalMolecularDynamics,
  MolecularMonteCarlo,
  CoarseGrainedMolecularSimulation,
  AbInitioMolecularDynamics,
  EnhancedSamplingMethods,
  FreeEnergySimulation,
  BiomolecularSimulation,
  ParallelHighPerformanceMolecularSimulation
 ]
 STOP_CONDITION = "Stop when representation, force field, ensemble, sampling method, numerical settings, equilibration, convergence, observables, validation and reproducibility are sufficiently resolved."
}

ROUTING {
 PARENT_ONTOLOGY = "Scientific Simulation Ontology"
 RELATED_ONTOLOGIES = [
  "Scientific Simulation Ontology",
  "Physical Simulation Ontology",
  "Particle Simulation Ontology",
  "Monte Carlo Simulation Ontology",
  "Random Process Ontology",
  "Numerical Methods Ontology 2",
  "Numerical Stability Ontology 2",
  "Numerical Error Ontology",
  "Floating Point Arithmetic Ontology",
  "Probability and Statistics Ontologies",
  "Scientific Computing Ontology 2"
 ]
 ROUTE_RULES = [
  "Molecular Simulation -> Scientific Simulation Ontology for parent semantics",
  "Molecular Simulation -> Physical Simulation for molecular physical-law context",
  "Molecular Simulation -> Particle Simulation for particle-state and neighbour semantics",
  "Molecular Simulation -> Monte Carlo Simulation for configurational sampling",
  "Molecular Simulation -> Random Process for Langevin and stochastic dynamics",
  "Molecular Simulation -> Numerical Methods for integration and sampling algorithms",
  "Molecular Simulation -> Numerical Stability for time-step and thermostat behaviour",
  "Molecular Simulation -> Numerical Error for cutoff, integration and statistical error",
  "Molecular Simulation -> Floating Point Arithmetic for accumulation and energy drift",
  "Molecular Simulation -> Probability and Statistics for ensembles and uncertainty",
  "Molecular Simulation -> Scientific Computing for scalable parallel execution"
 ]
}

UPDATE_MODEL {
 MODE = "append-compatible semantic versioning"
 MAJOR = "ontology-breaking conceptual change"
 MINOR = "new compatible force fields, sampling methods or reasoning pathways"
 PATCH = "clarification, metadata or formatting correction"
 RULES = [
  "Preserve molecular-representation-versus-force-field distinction.",
  "Preserve trajectory-dynamics-versus-Monte-Carlo-sampling distinction.",
  "Preserve equilibration-versus-production distinction.",
  "Preserve bonded-versus-nonbonded interaction distinction.",
  "Preserve short-range-versus-long-range treatment distinction.",
  "Preserve ensemble-definition-versus-control-algorithm distinction.",
  "Preserve raw-biased-sampling-versus-reweighted-observable distinction.",
  "Preserve machine-readable molecular-simulation semantics."
 ]
}

OUTPUT {
 CANONICAL_DEFINITION = "Molecular simulation computationally samples or evolves molecular configurations to study molecular-scale structure, dynamics and thermodynamics."
 REPRESENTATION_SUMMARY = "Atomistic, united-atom, coarse-grained and ab initio representations trade physical detail against computational cost."
 FORCE_FIELD_SUMMARY = "Bonded and nonbonded potentials define molecular interactions, while long-range methods treat electrostatics beyond local cutoffs."
 ENSEMBLE_SUMMARY = "Thermostats, barostats and Monte Carlo moves generate or approximate target statistical ensembles."
 SAMPLING_SUMMARY = "Enhanced-sampling and free-energy methods improve access to rare states and thermodynamic differences."
 CONVERGENCE_SUMMARY = "Autocorrelation, effective sample size, finite-size effects and time-step sensitivity determine statistical reliability."
 VALIDATION_SUMMARY = "Molecular simulations should be checked through energy behaviour, sampling convergence, structural and thermodynamic observables, reference data and reproducibility."
}

FLOW {
 INPUT
 -> detect_molecular_simulation_context
 -> identify_molecular_system_representation_thermodynamic_state_observables_accuracy_goal_and_budget
 -> define_topology_force_field_bonded_nonbonded_constraints_and_long_range_interactions
 -> choose_MD_MC_hybrid_enhanced_sampling_or_free_energy_method
 -> configure_ensemble_thermostat_barostat_time_step_cutoffs_neighbour_lists_and_long_range_solver
 -> initialise_coordinates_velocities_box_and_random_seed
 -> equilibrate_system
 -> execute_production_sampling
 -> collect_trajectory_and_observables
 -> assess_energy_drift_autocorrelation_effective_sample_size_finite_size_and_time_step_sensitivity
 -> if_enhanced_sampling_apply_reweighting_or_bias_correction
 -> verify_against_reference_calculations_and_conservation_checks
 -> validate_structural_thermodynamic_or_kinetic_observables_against_evidence
 -> document_force_field_versions_software_versions_configuration_seed_and_provenance
 -> route_to_related_scientific_simulation_physical_simulation_particle_simulation_Monte_Carlo_random_process_numerical_methods_stability_error_floating_point_probability_and_scientific_computing_ontologies
 -> recursively_expand_required_dependencies
 -> verify_result
 -> assign_confidence
 -> produce_semantic_output
 -> END
}

SEMANTIC_SUMMARY {
 "The Molecular Simulation Ontology defines computational modelling of molecular systems across classical molecular dynamics, molecular Monte Carlo, atomistic, united-atom, coarse-grained and ab initio representations. It supports molecular topology, force fields, bonded and nonbonded interactions, Lennard-Jones and Coulomb potentials, SHAKE and RATTLE constraints, Velocity-Verlet, leapfrog and multiple-time-step integration, NVE, NVT, NPT and grand-canonical ensembles, thermostats and barostats, Metropolis and configurational-bias Monte Carlo, replica exchange, metadynamics, umbrella sampling, adaptive biasing, free-energy perturbation, thermodynamic integration, BAR and MBAR, equilibration, autocorrelation and effective sample size, periodic boundaries, neighbour lists, Ewald, PME and P3M electrostatics, solvation models, biomolecular and materials applications, RDF, MSD and other observables, verification, experimental validation, uncertainty, parallel execution and reproducibility."
}

FLOW_HASH {
 METHOD = "semantic-content-identity"
 ID = "KIRETH-MATH-COMPUTATIONAL-MATHEMATICS-MOLECULAR-SIMULATION-ONTOLOGY-V1"
}

}

END_FLOW