Function reference • uqsa

Model and data

Functions for loading model and data and construct a mathematical model (ODE or stochastic) and corresponding code.

uqsa_example(): Load an example model for this package

model_from_tsv(): model_from_tsv loads the content from a series of tsv files

as_ode(): Interpret a model as an ODE

generateCode(): Write C code

write_and_compile(): Writes code to file and compiles

makeGillespieModel(): makeGillespieModel interprets the provided SBtab file as a stochastic model

generateGillespieCode(): Generate C Code to solve a model stochastically

generateRCode(): Write R code

values(): Find values in a data.frame that is derived from a tsv file or similar

experiments(): Extract Measured Data and Simulation Experiment Instructions

standard_error_matrix(): Standard Error Matrix from an errors object

checkModel(): checkModel tries to establish the simulation file for a given model

conservation_law_analysis(): Reduce the size of the system

parameterConversion(): Returns information about parameter conversion

replace_powers(): replace_powers

stoichiometry(): This function returns a list of named stoichiometric vectors

unit.from.string(): Unit Interpreter

unit.id(): Converts a unit to a string that works as an identifier

unit.info(): Prints an interpretation string of a unit

unit_as_character(): converts a unit data.frame into a printable string

units_from_table(): Get units from a data.frame column

yJacobian(): Jacobian of string-math

yacasMath(): yacasMath converts math to Ryacas compatible math

Simulations

Functions for simulating the model

simulator.c(): This creates a closure that simulates the model

simstoch(): Simulate stochastic model

simfi(): This creates a closure that simulates the model, similar to simulator.c

scrnn(): scrnn returns a closure around gsl_odeiv2_CRNN()

gsl_odeiv2_fi(): simulates an ode model with extra work

gsl_odeiv2_CRNN(): simulates a CRNN ode model with extra work

Prior

Functions for constructing priors

rNormalPrior(): rNormalPrior returns a random vector generator

rUniformPrior(): rUniformPrior returns a random vector generator

rCopulaPrior(): rCopulaPrior returns a function that generates random values from the copula model

dNormalPrior(): dNormalPrior creates the density function of a multivariate normal distribution with independent components

dUniformPrior(): dUniformPrior creates a uniform density function

dCopulaPrior(): dCopulaPrior creates a prior probability density function

gNormalPrior(): gNormalPrior creates the gradient function of a multivariate normal distribution with independent components, in log-space

fitCopula(): Makes a Probability Density Estimate (from a sample)

makeIndepCopula(): Copula Formulation for Uniform Prior Distributions

sampleWithNoise(): (for testing) A non-Copula sampling function as fallback

Uncertainty quantification - likelihood based

MCMC methods for Bayesian parameter inference

mcmc(): Markov Chain Monte Carlo

mcmc_init(): Initialize the Markov chain

mcmc_mpi(): The MPI version of the mcmc function

metropolis_update(): Metropolis Update is an MCMC update function

high_level_metropolis(): High Level Metropolis function

smmala_update(): SMMALA Update is an MCMC update function

smmala_move(): SMMALA move

smmala_move_density(): SMMALA transition kernel density

high_level_smmala(): High Level SMMALA function

logLikelihoodFunc(): Default log-likelihood function

gradLogLikelihoodFunc(): Default log-likelihood function, gradient

ll(): Default Log-likelihood Function

gllf(): Default gradient-log-likelihood Function

fi(): Default gradient-Log-likelihood Function

tune_step_size(): Find a good Step-Size for a given MCMC Algorithm

change_temperature(): Should 2 Markov chains exchange their temperatures

pbdMPI_bcast_reduce_temperatures(): Broadcast to other ranks and swap temperatures with any of them

loadSubSample_mpi(): This function merges mpi-samples into one

loadSample_mpi(): This function merges mpi-samples into one

gatherSample(): gatherSample collects all sample points, from all files, with the given temperature

gatherReplicas(): gatherReplicas collects all sample points, from all files, which are assumed to be exact replicas, with different seeds (and possibly sizes). This function uses mclapply to process the files, which may be quicker than gatherSample. The temperature is disregarded, assuming that no parallel tempering was used. To facilitate the loading of a very big sample, this function will analyse the auto-correltation within each file and returned a thinned subsample of size N/2*tauint (effective sample size). There is no need to further reduce the rfesult.

Uncertainty quantification - ABC

Approximate Bayesian Computation (ABC) MCMC for parameter inference

makeObjective(): creates Objective functions from ingredients

preCalibration(): Determine a starting value for ABC's delta

ABCMCMC(): Performs and Approximate Bayesian Computation Sampling of Model Parameters

checkFitWithPreviousExperiments(): ABC acceptance of currently sampled values given old data (Prior)

Global Sensitivity Analysis

gsa_binning(): Global Sensitivity Analysis

gsa_saltelli(): Outputs the global sensitivity scores SI and SIT, calculated by the Sobol-Homma-Saltelli method

saltelli_prior(): Sample for the Sobol-Homma-Saltelli Global Sensitivity Analysis

Parameter mappings

Parameter transformations from sampling space back to model space

logParMap(): NATURAL LOG parameter mapping used by the MCMC module

logParMapJac(): NATURAL LOG parameter mapping, jacobian

log10ParMap(): LOG10 parameter mapping used by the MCMC module

log10ParMapJac(): LOG10 parameter mapping, jacobian

log2ParMap(): LOG2 parameter mapping used by the MCMC module

log2ParMapJac(): LOG2 parameter mapping, jacobian

Operators and helper functions

`%@%`: Fetch an Attribute

`%as%`: %as% is a binary operator on strings with units in them

`%has%`: checks whether a variable has the named attributes

`%otherwise%`: This function can be used to specify default values

determinePrefix(): Determine a prefix from a character vector str of similar contents

Plotting functions

pcDist(): plots a sample in parallel coordinates

plotTimeSeries(): Plot time series simulations with experimental data

plotTimeSeriesBase(): plot function for experiments

sensitivity.graph(): plot the sensitivity matrix

showPosterior(): showPosterior makes a pairs plot for a sample

Internal functions

`modify<-`(): Modifies a value

getDose(): Get the values of the input for a series of dose response experiments

ggplotDoseResponse(): Plot dose response simulations with experimental data

ggplotTimeSeries(): Plot time series simulations with experimental data

ggplotTimeSeriesStates(): Plot time series simulation with state variables

onlyCoefficients(): Returns a list of reaction coefficients

onlyNames(): Returns only the names in a reaction formula

parUpdate(): Updates Parameter Values

simfiGaussianLogLikelihood(): SMMALA -- The default Extractor of the log-likelihood computed by the simfi solver

CRNN(): CRNN creates C code for a chemical reaction neural network

automatic_update(): This function proposes an MCMC candidate variable, and either accepts or rejects the candidate

`base<-`(): Convert to linear space

column(): Get j-th column with names

defaultAcceptance(): default ABC acceptance probability function for one experiment

defaultDistance(): default distance function for one experiment

find_parameter_reaction(): Find the stoichiometry for a given parameter name

flux_matrix(): Creates a Matrix with reaction kinetic entries

formulae(): Find a column that contains some kind of mathematic expression in a data.frame

getMCMCPar(): Selects MCMC scheme specific setup parameters

initialCount(): Initial count of reacting Compounds

is.invertible(): checks whether a given matrix is a valid, invertible fisherInformation

linear_scale(): Interprets a character vector as names of logarithms

moleCountConversion(): This function calculates the conversion from moles to particle counts

observable.mean.in.bin(): The mean value of an observable value given a parameter bin

parameterReaction(): Find the Reaction a parameter appears in

parse_concise(): Read Concise Error Notation

`reaction<-`(): Add a Reaction to an ODE

update_values(): Updates the named values of vector v with values mentioned in data.frame d

writeCFunction(): Write C Function

scaleParameter(): Returns a string that contains code to scale the parameter

simple.unit(): Simple unit from string

sum.of.bin.variance(): Weighted Sum of Bin-specific variances

unit.kind(): find unit category

unit.scale(): Unit scale from SI prefix