simBsOpt: R6 Class Representing a Breeding Scheme
In KosukeHamazaki/myBreedSimulatR: R Breeding Simulator for K.Hamazaki
simBsOpt R Documentation
R6 Class Representing a Breeding Scheme
Description
simBsOpt object store specific information of simulation results of breeding scheme.
Public fields
simBsName[character] Name of this simulation of breeding schemes
bsInfoInit[bsInfo] breeding scheme info
(see:bsInfo)
breederInfoInit[breederInfo] breeder info
(see:breederInfo)
lociEffMethod[character] How to compute loci effects (use true QTL effects ("true") or estimated marker effects ("estimated"))
trainingPopType[character] Training population consists of all populations created in the breeding scheme for 'trainingPopType = "all"',
or consists of the latest population in breeding scheme for 'trainingPopType = "latest"'.
trainingPopInit[character / numeric] Training population names or No.s (not generations!!) for initial population
trainingIndNamesInit[character] Names of training individuals for initia; population
methodMLRInit[character] Methods for estimating marker effects for initial population.
The following methods are offered:
"Ridge", "LASSO", "ElasticNet", "RR-BLUP", "GBLUP", "BayesA" (uni-trait),
"BayesB" (uni-trait), "BayesC" (uni-trait), "BRR", "BL" (uni-trait), "SpikeSlab" (multi-trait)
multiTraitInit[logical] Use multiple-trait model for estimation of marker effects or not for initial population
nIterSimulation[numeric] Number of iterations for this simulation setting
nGenerationProceed[numeric] Number of generations to be proceeded
nGenerationProceedSimulation[numeric] Number of generations to be proceeded in simulation for optimizing policy
setGoalAsFinalGeneration[logical] Set goal for simulation of breeding scheme as the real final generation
performOptimization[logical] Perform optimization of policy in each generation or not
useFirstOptimizedValue[logical] Perform optimization of policy only for the initial population and use the optimized hyperprameters permanently.
sameAcrossGeneration[logical] Use same hyper prameter across generations or not
hMin[numeric] Lower bound of hyper parameters
hMax[numeric] Upper bound of hyper parameters
nTotalIterForOneOptimization[numeric] Number of total iterations that can be assigned for one optimization process
nIterSimulationPerEvaluation[numeric] Number of simulations per one evaluation of the hyperparameter set of your interest
nIterOptimization[numeric] Number of iterations required for one optimization
nMaxEvalPerNode[numeric] Number of maximum evaluation per node when optimizing hyper parameter by StoSOO
maxDepth[numeric] Maximum depth of tree when optimizing hyper parameter by StoSOO
nChildrenPerExpansion[numeric] Number of children per one expansion of nodes when optimizing hyper parameter by StoSOO
confidenceParam[numeric] Confidence parameter of StoSOO, this parameter determines the width of the estimates of rewards
returnOptimalNodes[numeric] When (how many iterations) to return (or save) the optimal nodes when optimizing hyper parameter by StoSOO
saveTreeNameBase[character] Base name of the tree to be saved
whenToSaveTrees[numeric] When (how many iterations) to save the tree in StoSOO
nTopEvalForOpt[numeric] Number of individuals to be evaluated when evaluating population max or population min for optimization of hyperparameters
rewardWeightVec[numeric] When returning reward function, 'rewardWeightVec' will be multiplied by estimated GVs for each generation to evaluate the method.
If you want to apply discounted method, you can achieve by 'rewardWeightVec = sapply(1:nGenerationProceedSimulation, function(genProceedNo) gamma ^ (genProceedNo - 1))' where 'gamma' is discounted rate.
Or if you want to evaluate just the final generation, you can achieve by 'rewardWeightVec = c(rep(0, nGenerationProceedSimulation - 1), 1)'.
digitsEval[numeric] When you evaluate each hyperparameter, you can round the average of evaluates ('eval') with 'round(eval, digitsEval)'.
nRefreshMemoryEvery[numeric] Every 'nRefreshMemoryEvery' iterations, we refresh memory used for simulations by 'gc(reset = TRUE)'.
updateBreederInfo[logical] Update breederInfo or not for each generation
phenotypingInds[logical] Phenotyping individuals or not for each generation
nRepForPhenoInit[numeric] Number of replications to be phenotyped for initial population
nRepForPheno[numeric] Number of replications to be phenotyped for each generation
updateModels[logical] Whether or not updating the model for each generation
(If 'phenotypingInds = FALSE' for generation of your interest, 'updateModels' will be automatically 'FALSE' for that generation.)
updateBreederInfoSimulation[logical] Update breederInfo or not for each generation in simulations for optimization
phenotypingIndsSimulation[logical] Phenotyping individuals or not for each generation in simulations for optimization
nRepForPhenoSimulation[numeric] Number of replications to be phenotyped for each generation in simulations for optimization
updateModelsSimulation[logical] Whether or not updating the model for each generation in simulations for optimization
(If 'phenotypingInds = FALSE' for generation of your interest, 'updateModels' will be automatically 'FALSE' for that generation.)
methodMLR[character] Methods for estimating marker effects.
The following methods are offered:
"Ridge", "LASSO", "ElasticNet", "RR-BLUP", "GBLUP", "BayesA" (uni-trait),
"BayesB" (uni-trait), "BayesC" (uni-trait), "BRR", "BL" (uni-trait), "SpikeSlab" (multi-trait)
multiTrait[logical] Use multiple-trait model for estimation of marker effects or not
nSelectionWaysVec[numeric] Number of selection ways for each generation
selectionMethodList[list] (list of) Selection method for each generation
traitNoSelList[list] (list of list of) Number of trait No of your interest for selection for each generation
blockSplitMethod[character] How to determine the markers belonging to each block when computing OHV.
You can define the number of markers in each block ('nMrkInBlock') or the minimum length of each segment ('minimumSegmentLength').
nMrkInBlock[numeric] Number of markers in each block. This will be used for the computation of OHV.
minimumSegmentLength[numeric] Minimum length of each segment [cM]. This will be used for the computation of OHV.
nSelInitOPVList[list] (list of) Number of selected candiates for first screening before selecting parent candidates by OPV for each generation
nIterOPV[numeric] Number of iterations for computation of OPV
nProgeniesEMBVVec[numeric] Number of progenies of double haploids produced when computing EMBV for each generation
nIterEMBV[numeric] Number of iterations to estimate EMBV
nCoresEMBV[numeric] Number of cores used for EMBV estimation
clusteringForSelList[list] (list of) Apply clustering results of marker genotype to selection or not for each generation
nClusterList[list] (list of) Number of clusters for each generation
nTopClusterList[list] (list of) Number of top clusters used for selection for each generation
nTopEachList[list] (list of) Number of selected individuals in each cluster for each generation
nSelList[list] (list of) Number of selection candidates for each generation
multiTraitsEvalMethodList[list] (list of) When evaluating multiple traits, you can choose how to evaluate these traits simultaneously.
One method is to take a weighted sum of these traits, and the other is to compute a product of these traits (adjusted to be positive values in advance.)
hSelList[list] (list of) Hyperparameter which determines which trait is weighted when selecting parent candidates for multiple traits for each generation
matingMethodVec[character] Mating method for each generation
allocateMethodVec[character] Allocation method for each generation
weightedAllocationMethodList[list] (list of) Which selection index will be used for weighted resource allocation for each generation
traitNoRAList[list] (list of) Trait No of your interest for resource allocation for each generation
includeGVPVec[logical] Whether or not to consider genetic variance of progenies of each pair when determining the number of progenies per each pair for each generation
nNextPopVec[numeric] Number of progenies for the next generation for each generation
nameMethod[character] Method for naming individuals
nCores[numeric] Number of cores used for simulations of breedinhh scheme
nCoresPerOptimization[numeric] Number of cores used for simulations of breeding scheme when optimizing hyperparameters
overWriteRes[logical] Overwrite simulation results when the targeted results already exists
showProgress[logical] Show progress bar or not
returnMethod[character] Which type of results will be returned (saved) in the object
saveAllResAt[character] If NULL, we won't save the simulation results. Else, we will save bsInfo and breederInfo for each iteration in the directory defined by 'saveAllResAt' path.
evaluateGVMethod[character] Which type of GV (true GV or estimated GV) will be used for evaluation of the simulation results.
nTopEval[numeric] Number of individuals to be evaluated when evaluating population max or population min
traitNoEval[numeric] Trait No of your interest for evaluation of the method
hEval[numeric] Hyperparameter which determines which BV is emphasized when evaluating simulation results of each method
summaryAllResAt[character] If NULL, we will summary the simulation results in 'self$trueGVMatList' and 'self$estimatedGVMatList'.
Else, we will summary the simulation results in 'summaryAllResAt' path.
verbose[logical] Display info (optional)
lociEffectsInit[matrix] Marker and QTL effects used for crossInfo object for initial population
hLens[numeric] Length of hyperparameter vector
hStart[numeric] Initial values of hyper parameters
solnInit[list] List of solution of StoSOO for initial state
hVecOptsList[list] List of optimized hyperparameters in each generation
optimalHyperParamMatsList[list] List of optimized hyper parameters given finite numbers of budget for optimization
simBsRes[list] Simulation results of each method
trueGVMatInit[matrix] A true GV matrix for initial population
trueGVMatList[list] A list of true GV matrix of simulation results
trueGVSummaryArray[array] An array of summary statistics of true GVs for each population & each iteration
estimatedGVMatInit[matrix] An estimated GV matrix for initial population
estimatedGVMatList[list] A list of estimated GV matrix of simulation results
estimatedGVSummaryArray[array] An array of summary statistics of estimated GVs for each population & each iteration
Methods
Public methods
-
-
-
-
-
-
-
Method new()
Create a new simBsOpt object.
Usage
simBsOpt$new(
simBsName = "Undefined",
bsInfoInit,
breederInfoInit = NULL,
lociEffMethod = NULL,
trainingPopType = NULL,
trainingPopInit = NULL,
trainingIndNamesInit = NULL,
methodMLRInit = NULL,
multiTraitInit = FALSE,
nIterSimulation = NULL,
nGenerationProceed = NULL,
nGenerationProceedSimulation = NULL,
setGoalAsFinalGeneration = TRUE,
performOptimization = NULL,
useFirstOptimizedValue = NULL,
sameAcrossGeneration = TRUE,
hMin = NULL,
hMax = NULL,
nTotalIterForOneOptimization = NULL,
nIterSimulationPerEvaluation = NULL,
nIterOptimization = NULL,
nMaxEvalPerNode = NULL,
maxDepth = NULL,
nChildrenPerExpansion = NULL,
confidenceParam = NULL,
returnOptimalNodes = NULL,
saveTreeNameBase = NULL,
whenToSaveTrees = NA,
nTopEvalForOpt = NULL,
rewardWeightVec = NULL,
digitsEval = NULL,
nRefreshMemoryEvery = NULL,
updateBreederInfo = TRUE,
phenotypingInds = FALSE,
nRepForPhenoInit = NULL,
nRepForPheno = NULL,
updateModels = FALSE,
updateBreederInfoSimulation = TRUE,
phenotypingIndsSimulation = FALSE,
nRepForPhenoSimulation = NULL,
updateModelsSimulation = FALSE,
methodMLR = NULL,
multiTrait = FALSE,
nSelectionWaysVec = NULL,
selectionMethodList = NULL,
traitNoSelList = NULL,
blockSplitMethod = NULL,
nMrkInBlock = NULL,
minimumSegmentLength = NULL,
nSelInitOPVList = NULL,
nIterOPV = NULL,
nProgeniesEMBVVec = NULL,
nIterEMBV = NULL,
nCoresEMBV = NULL,
clusteringForSelList = FALSE,
nClusterList = NULL,
nTopClusterList = NULL,
nTopEachList = NULL,
nSelList = NULL,
multiTraitsEvalMethodList = NULL,
hSelList = NULL,
matingMethodVec = NULL,
allocateMethodVec = NULL,
weightedAllocationMethodList = NULL,
traitNoRAList = NULL,
includeGVPVec = FALSE,
nNextPopVec = NULL,
nameMethod = "pairBase",
nCores = NULL,
nCoresPerOptimization = NULL,
overWriteRes = FALSE,
showProgress = TRUE,
returnMethod = "all",
saveAllResAt = NULL,
evaluateGVMethod = "true",
nTopEval = NULL,
traitNoEval = NULL,
hEval = NULL,
summaryAllResAt = NULL,
verbose = TRUE
)
Arguments
simBsName[character] Name of this simulation of breeding schemes
bsInfoInit[bsInfo] breeding scheme info
(see:bsInfo)
breederInfoInit[breederInfo] breeder info
(see:breederInfo)
lociEffMethod[character] How to compute loci effects (use true QTL effects ("true") or estimated marker effects ("estimated"))
trainingPopType[character] Training population consists of all populations created in the breeding scheme for 'trainingPopType = "all"',
or consists of the latest population in breeding scheme for 'trainingPopType = "latest"'.
trainingPopInit[character / numeric] Training population names or No.s (not generations!!) for initial population
trainingIndNamesInit[character] Names of training individuals for initia; population
methodMLRInit[character] Methods for estimating marker effects for initial population.
The following methods are offered:
"Ridge", "LASSO", "ElasticNet", "RR-BLUP", "GBLUP", "BayesA" (uni-trait),
"BayesB" (uni-trait), "BayesC" (uni-trait), "BRR", "BL" (uni-trait), "SpikeSlab" (multi-trait)
multiTraitInit[logical] Use multiple-trait model for estimation of marker effects or not for initial population
nIterSimulation[numeric] Number of iterations for this simulation setting
nGenerationProceed[numeric] Number of generations to be proceeded
nGenerationProceedSimulation[numeric] Number of generations to be proceeded in simulation for optimizing policy
setGoalAsFinalGeneration[logical] Set goal for simulation of breeding scheme as the real final generation
performOptimization[logical] Perform optimization of policy in each generation or not
useFirstOptimizedValue[logical] Perform optimization of policy only for the initial population and use the optimized hyperprameters permanently.
sameAcrossGeneration[logical] Use same hyper prameter across generations or not
hMin[numeric] Lower bound of hyper parameters
hMax[numeric] Upper bound of hyper parameters
nTotalIterForOneOptimization[numeric] Number of total iterations that can be assigned for one optimization process
nIterSimulationPerEvaluation[numeric] Number of simulations per one evaluation of the hyperparameter set of your interest
nIterOptimization[numeric] Number of iterations required for one optimization
nMaxEvalPerNode[numeric] Number of maximum evaluation per node when optimizing hyper parameter by StoSOO
maxDepth[numeric] Maximum depth of tree when optimizing hyper parameter by StoSOO
nChildrenPerExpansion[numeric] Number of children per one expansion of nodes when optimizing hyper parameter by StoSOO
confidenceParam[numeric] Confidence parameter of StoSOO, this parameter determines the width of the estimates of rewards
returnOptimalNodes[numeric] When (how many iterations) to return (or save) the optimal nodes when optimizing hyper parameter by StoSOO
saveTreeNameBase[character] Base name of the tree to be saved
whenToSaveTrees[numeric] When (how many iterations) to save the tree in StoSOO
nTopEvalForOpt[numeric] Number of individuals to be evaluated when evaluating population max or population min for optimization of hyperparameters
rewardWeightVec[numeric] When returning reward function, 'rewardWeightVec' will be multiplied by estimated GVs for each generation to evaluate the method.
If you want to apply discounted method, you can achieve by 'rewardWeightVec = sapply(1:nGenerationProceedSimulation, function(genProceedNo) gamma ^ (genProceedNo - 1))' where 'gamma' is discounted rate.
Or if you want to evaluate just the final generation, you can achieve by 'rewardWeightVec = c(rep(0, nGenerationProceedSimulation - 1), 1)'.
digitsEval[numeric] When you evaluate each hyperparameter, you can round the average of evaluates ('eval') with 'round(eval, digitsEval)'.
nRefreshMemoryEvery[numeric] Every 'nRefreshMemoryEvery' iterations, we refresh memory used for simulations by 'gc(reset = TRUE)'.
updateBreederInfo[logical] Update breederInfo or not for each generation
phenotypingInds[logical] Phenotyping individuals or not for each generation
nRepForPhenoInit[numeric] Number of replications to be phenotyped for initial population
nRepForPheno[numeric] Number of replications to be phenotyped for each generation
updateModels[logical] Whether or not updating the model for each generation
(If 'phenotypingInds = FALSE' for generation of your interest, 'updateModels' will be automatically 'FALSE' for that generation.)
updateBreederInfoSimulation[logical] Update breederInfo or not for each generation in simulations for optimization
phenotypingIndsSimulation[logical] Phenotyping individuals or not for each generation in simulations for optimization
nRepForPhenoSimulation[numeric] Number of replications to be phenotyped for each generation in simulations for optimization
updateModelsSimulation[logical] Whether or not updating the model for each generation in simulations for optimization
(If 'phenotypingInds = FALSE' for generation of your interest, 'updateModels' will be automatically 'FALSE' for that generation.)
methodMLR[character] Methods for estimating marker effects.
The following methods are offered:
"Ridge", "LASSO", "ElasticNet", "RR-BLUP", "GBLUP", "BayesA" (uni-trait),
"BayesB" (uni-trait), "BayesC" (uni-trait), "BRR", "BL" (uni-trait), "SpikeSlab" (multi-trait)
multiTrait[logical] Use multiple-trait model for estimation of marker effects or not
nSelectionWaysVec[numeric] Number of selection ways for each generation
selectionMethodList[list] (list of) Selection method for each generation
traitNoSelList[list] (list of list of) Number of trait No of your interest for selection for each generation
blockSplitMethod[character] How to determine the markers belonging to each block when computing OHV.
You can define the number of markers in each block ('nMrkInBlock') or the minimum length of each segment ('minimumSegmentLength').
nMrkInBlock[numeric] Number of markers in each block. This will be used for the computation of OHV.
minimumSegmentLength[numeric] Minimum length of each segment [cM]. This will be used for the computation of OHV.
nSelInitOPVList[list] (list of) Number of selected candiates for first screening before selecting parent candidates by OPV for each generation
nIterOPV[numeric] Number of iterations for computation of OPV
nProgeniesEMBVVec[numeric] Number of progenies of double haploids produced when computing EMBV for each generation
nIterEMBV[numeric] Number of iterations to estimate EMBV
nCoresEMBV[numeric] Number of cores used for EMBV estimation
clusteringForSelList[list] (list of) Apply clustering results of marker genotype to selection or not for each generation
nClusterList[list] (list of) Number of clusters for each generation
nTopClusterList[list] (list of) Number of top clusters used for selection for each generation
nTopEachList[list] (list of) Number of selected individuals in each cluster for each generation
nSelList[list] (list of) Number of selection candidates for each generation
multiTraitsEvalMethodList[list] (list of) When evaluating multiple traits, you can choose how to evaluate these traits simultaneously.
One method is to take a weighted sum of these traits, and the other is to compute a product of these traits (adjusted to be positive values in advance.)
hSelList[list] (list of) Hyperparameter which determines which trait is weighted when selecting parent candidates for multiple traits for each generation
matingMethodVec[character] Mating method for each generation
allocateMethodVec[character] Allocation method for each generation
weightedAllocationMethodList[list] (list of) Which selection index will be used for weighted resource allocation for each generation
traitNoRAList[list] (list of) Trait No of your interest for resource allocation for each generation
includeGVPVec[logical] Whether or not to consider genetic variance of progenies of each pair when determining the number of progenies per each pair for each generation
nNextPopVec[numeric] Number of progenies for the next generation for each generation
nameMethod[character] Method for naming individuals
nCores[numeric] Number of cores used for simulations of breeding scheme
nCoresPerOptimization[numeric] Number of cores used for simulations of breeding scheme when optimizing hyperparameters
overWriteRes[logical] Overwrite simulation results when the targeted results already exists
showProgress[logical] Show progress bar or not
returnMethod[character] Which type of results will be returned (saved) in the object
saveAllResAt[character] If NULL, we won't save the simulation results. Else, we will save bsInfo and breederInfo for each iteration in the directory defined by 'saveAllResAt' path.
evaluateGVMethod[character] Which type of GV (true GV or estimated GV) will be used for evaluation of the simulation results
nTopEval[numeric] Number of individuals to be evaluated when evaluating population max or population min#'
traitNoEval[numeric] Trait No of your interest for evaluation of the method
hEval[numeric] Hyperparameter which determines which BV is emphasized when evaluating simulation results of each method
summaryAllResAt[character] If NULL, we will summary the simulation results in 'self$trueGVMatList' and 'self$estimatedGVMatList'.
Else, we will summary the simulation results in 'summaryAllResAt' path.
verbose[logical] Display info (optional)
Returns
A new 'simBs' object.
Examples
### create simulation information
mySimInfo <- simInfo$new(simName = "Simulation Example",
simGeno = TRUE,
simPheno = TRUE,
#' nSimGeno = 1,
#' nSimPheno = 3,
#' nCoreMax = 4,
#' nCorePerGeno = 1,
#' nCorePerPheno = 3,
saveDataFileName = NULL)
### create specie information
mySpec <- specie$new(nChr = 3,
lChr = c(100, 150, 200),
specName = "Example 1",
ploidy = 2,
mutRate = 10^-8,
recombRate = 10^-6,
chrNames = c("C1", "C2", "C3"),
nLoci = 100,
recombRateOneVal = FALSE,
effPopSize = 100,
simInfo = mySimInfo,
verbose = TRUE)
### create lociInfo object
myLoci <- lociInfo$new(genoMap = NULL, specie = mySpec)
plot(myLoci, alpha = 0.1)
### create traitInfo object
myTrait <- traitInfo$new(lociInfo = myLoci,
nMarkers = 80,
nTraits = 3,
nQTLs = c(4, 8, 3),
actionTypeEpiSimple = TRUE,
qtlOverlap = TRUE,
nOverlap = c(2, 3, 0),
effCor = 0.1,
propDomi = 0.2,
interactionMean = c(4, 1, 2))
myTrait$plot(alphaMarker = 0.1)
### create bsInfo object
myBsInfo <- bsInfo$new(simInfo = mySimInfo,
specie = mySpec,
lociInfo = myLoci,
traitInfo = myTrait,
geno = NULL,
haplo = NULL,
founderIsInitPop = TRUE,
popNameBase = "Population",
initIndNames = NULL,
verbose = TRUE)
### create cross information object
for (i in 1:10) {
myCrossInfo <- crossInfo$new(parentPopulation = myBsInfo$populations[[myBsInfo$generation]],
method = "randomMate",
nNextPop = 100)
myBsInfo$nextGeneration(crossInfo = myCrossInfo)
}
geno <- myBsInfo$overGeneration()$genoMat
myBsInfo$print()
myBsInfo$plot(plotTarget = "trueAGV",
targetTrait = 1:3,
targetPopulation = 1:11,
plotType = "jitter")
Method computeLociEffInit()
estimate/extract marker/QTL effects information
Usage
simBsOpt$computeLociEffInit()
Method startSimulation()
start simulation of breeding scheme
Usage
simBsOpt$startSimulation()
Method summaryResults()
start simulation of breeding scheme
Usage
simBsOpt$summaryResults()
Method print()
Display information about the object
Usage
simBsOpt$print()
Method plot()
Draw figures for visualization of simulation results for summary statistics
Usage
simBsOpt$plot(
targetTrait = 1,
targetPopulation = NULL,
plotType = "box",
plotTargetDensity = "max",
plotGVMethod = "true",
adjust = 1e-05
)
Arguments
targetTrait[numeric] Target trait. character is OK, but numeric vector
corresponding to target traits is preferred. It should be a vector with length 1.
targetPopulation[numeric] Target populations. character is OK, but numeric vector
corresponding to target traits is preferred.
plotType[character] We offer "box", "violin", "lines", "density"
to draw figures for simulation results.
plotTargetDensity[character] When you choose "density" for 'plotType',
you should select which summary statistics will be plotted. It should be a vector with length 1.
plotGVMethod[character] Which type of GV (true GV or estimated GV) will be used for plotting the simulation results
adjust[numeric] the bandwidth used is actually adjust*bw. This makes it easy to specify values like ‘half the default’ bandwidth.
(see: 'adjust' in density)
Method clone()
The objects of this class are cloneable with this method.
Usage
simBsOpt$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.
Examples
## ------------------------------------------------
## Method `simBsOpt$new`
## ------------------------------------------------
### create simulation information
mySimInfo <- simInfo$new(simName = "Simulation Example",
simGeno = TRUE,
simPheno = TRUE,
#' nSimGeno = 1,
#' nSimPheno = 3,
#' nCoreMax = 4,
#' nCorePerGeno = 1,
#' nCorePerPheno = 3,
saveDataFileName = NULL)
### create specie information
mySpec <- specie$new(nChr = 3,
lChr = c(100, 150, 200),
specName = "Example 1",
ploidy = 2,
mutRate = 10^-8,
recombRate = 10^-6,
chrNames = c("C1", "C2", "C3"),
nLoci = 100,
recombRateOneVal = FALSE,
effPopSize = 100,
simInfo = mySimInfo,
verbose = TRUE)
### create lociInfo object
myLoci <- lociInfo$new(genoMap = NULL, specie = mySpec)
plot(myLoci, alpha = 0.1)
### create traitInfo object
myTrait <- traitInfo$new(lociInfo = myLoci,
nMarkers = 80,
nTraits = 3,
nQTLs = c(4, 8, 3),
actionTypeEpiSimple = TRUE,
qtlOverlap = TRUE,
nOverlap = c(2, 3, 0),
effCor = 0.1,
propDomi = 0.2,
interactionMean = c(4, 1, 2))
myTrait$plot(alphaMarker = 0.1)
### create bsInfo object
myBsInfo <- bsInfo$new(simInfo = mySimInfo,
specie = mySpec,
lociInfo = myLoci,
traitInfo = myTrait,
geno = NULL,
haplo = NULL,
founderIsInitPop = TRUE,
popNameBase = "Population",
initIndNames = NULL,
verbose = TRUE)
### create cross information object
for (i in 1:10) {
myCrossInfo <- crossInfo$new(parentPopulation = myBsInfo$populations[[myBsInfo$generation]],
method = "randomMate",
nNextPop = 100)
myBsInfo$nextGeneration(crossInfo = myCrossInfo)
}
geno <- myBsInfo$overGeneration()$genoMat
myBsInfo$print()
myBsInfo$plot(plotTarget = "trueAGV",
targetTrait = 1:3,
targetPopulation = 1:11,
plotType = "jitter")
KosukeHamazaki/myBreedSimulatR documentation built on Aug. 31, 2024, 3:55 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
| simBsOpt | R Documentation |
R6 Class Representing a Breeding Scheme
Description
simBsOpt object store specific information of simulation results of breeding scheme.
Public fields
simBsName[character] Name of this simulation of breeding schemes
bsInfoInit[bsInfo] breeding scheme info (see:bsInfo)
breederInfoInit[breederInfo] breeder info (see:breederInfo)
lociEffMethod[character] How to compute loci effects (use true QTL effects ("true") or estimated marker effects ("estimated"))
trainingPopType[character] Training population consists of all populations created in the breeding scheme for 'trainingPopType = "all"', or consists of the latest population in breeding scheme for 'trainingPopType = "latest"'.
trainingPopInit[character / numeric] Training population names or No.s (not generations!!) for initial population
trainingIndNamesInit[character] Names of training individuals for initia; population
methodMLRInit[character] Methods for estimating marker effects for initial population. The following methods are offered:
"Ridge", "LASSO", "ElasticNet", "RR-BLUP", "GBLUP", "BayesA" (uni-trait), "BayesB" (uni-trait), "BayesC" (uni-trait), "BRR", "BL" (uni-trait), "SpikeSlab" (multi-trait)
multiTraitInit[logical] Use multiple-trait model for estimation of marker effects or not for initial population
nIterSimulation[numeric] Number of iterations for this simulation setting
nGenerationProceed[numeric] Number of generations to be proceeded
nGenerationProceedSimulation[numeric] Number of generations to be proceeded in simulation for optimizing policy
setGoalAsFinalGeneration[logical] Set goal for simulation of breeding scheme as the real final generation
performOptimization[logical] Perform optimization of policy in each generation or not
useFirstOptimizedValue[logical] Perform optimization of policy only for the initial population and use the optimized hyperprameters permanently.
sameAcrossGeneration[logical] Use same hyper prameter across generations or not
hMin[numeric] Lower bound of hyper parameters
hMax[numeric] Upper bound of hyper parameters
nTotalIterForOneOptimization[numeric] Number of total iterations that can be assigned for one optimization process
nIterSimulationPerEvaluation[numeric] Number of simulations per one evaluation of the hyperparameter set of your interest
nIterOptimization[numeric] Number of iterations required for one optimization
nMaxEvalPerNode[numeric] Number of maximum evaluation per node when optimizing hyper parameter by StoSOO
maxDepth[numeric] Maximum depth of tree when optimizing hyper parameter by StoSOO
nChildrenPerExpansion[numeric] Number of children per one expansion of nodes when optimizing hyper parameter by StoSOO
confidenceParam[numeric] Confidence parameter of StoSOO, this parameter determines the width of the estimates of rewards
returnOptimalNodes[numeric] When (how many iterations) to return (or save) the optimal nodes when optimizing hyper parameter by StoSOO
saveTreeNameBase[character] Base name of the tree to be saved
whenToSaveTrees[numeric] When (how many iterations) to save the tree in StoSOO
nTopEvalForOpt[numeric] Number of individuals to be evaluated when evaluating population max or population min for optimization of hyperparameters
rewardWeightVec[numeric] When returning reward function, 'rewardWeightVec' will be multiplied by estimated GVs for each generation to evaluate the method. If you want to apply discounted method, you can achieve by 'rewardWeightVec = sapply(1:nGenerationProceedSimulation, function(genProceedNo) gamma ^ (genProceedNo - 1))' where 'gamma' is discounted rate. Or if you want to evaluate just the final generation, you can achieve by 'rewardWeightVec = c(rep(0, nGenerationProceedSimulation - 1), 1)'.
digitsEval[numeric] When you evaluate each hyperparameter, you can round the average of evaluates ('eval') with 'round(eval, digitsEval)'.
nRefreshMemoryEvery[numeric] Every 'nRefreshMemoryEvery' iterations, we refresh memory used for simulations by 'gc(reset = TRUE)'.
updateBreederInfo[logical] Update breederInfo or not for each generation
phenotypingInds[logical] Phenotyping individuals or not for each generation
nRepForPhenoInit[numeric] Number of replications to be phenotyped for initial population
nRepForPheno[numeric] Number of replications to be phenotyped for each generation
updateModels[logical] Whether or not updating the model for each generation (If 'phenotypingInds = FALSE' for generation of your interest, 'updateModels' will be automatically 'FALSE' for that generation.)
updateBreederInfoSimulation[logical] Update breederInfo or not for each generation in simulations for optimization
phenotypingIndsSimulation[logical] Phenotyping individuals or not for each generation in simulations for optimization
nRepForPhenoSimulation[numeric] Number of replications to be phenotyped for each generation in simulations for optimization
updateModelsSimulation[logical] Whether or not updating the model for each generation in simulations for optimization (If 'phenotypingInds = FALSE' for generation of your interest, 'updateModels' will be automatically 'FALSE' for that generation.)
methodMLR[character] Methods for estimating marker effects. The following methods are offered:
"Ridge", "LASSO", "ElasticNet", "RR-BLUP", "GBLUP", "BayesA" (uni-trait), "BayesB" (uni-trait), "BayesC" (uni-trait), "BRR", "BL" (uni-trait), "SpikeSlab" (multi-trait)
multiTrait[logical] Use multiple-trait model for estimation of marker effects or not
nSelectionWaysVec[numeric] Number of selection ways for each generation
selectionMethodList[list] (list of) Selection method for each generation
traitNoSelList[list] (list of list of) Number of trait No of your interest for selection for each generation
blockSplitMethod[character] How to determine the markers belonging to each block when computing OHV. You can define the number of markers in each block ('nMrkInBlock') or the minimum length of each segment ('minimumSegmentLength').
nMrkInBlock[numeric] Number of markers in each block. This will be used for the computation of OHV.
minimumSegmentLength[numeric] Minimum length of each segment [cM]. This will be used for the computation of OHV.
nSelInitOPVList[list] (list of) Number of selected candiates for first screening before selecting parent candidates by OPV for each generation
nIterOPV[numeric] Number of iterations for computation of OPV
nProgeniesEMBVVec[numeric] Number of progenies of double haploids produced when computing EMBV for each generation
nIterEMBV[numeric] Number of iterations to estimate EMBV
nCoresEMBV[numeric] Number of cores used for EMBV estimation
clusteringForSelList[list] (list of) Apply clustering results of marker genotype to selection or not for each generation
nClusterList[list] (list of) Number of clusters for each generation
nTopClusterList[list] (list of) Number of top clusters used for selection for each generation
nTopEachList[list] (list of) Number of selected individuals in each cluster for each generation
nSelList[list] (list of) Number of selection candidates for each generation
multiTraitsEvalMethodList[list] (list of) When evaluating multiple traits, you can choose how to evaluate these traits simultaneously. One method is to take a weighted sum of these traits, and the other is to compute a product of these traits (adjusted to be positive values in advance.)
hSelList[list] (list of) Hyperparameter which determines which trait is weighted when selecting parent candidates for multiple traits for each generation
matingMethodVec[character] Mating method for each generation
allocateMethodVec[character] Allocation method for each generation
weightedAllocationMethodList[list] (list of) Which selection index will be used for weighted resource allocation for each generation
traitNoRAList[list] (list of) Trait No of your interest for resource allocation for each generation
includeGVPVec[logical] Whether or not to consider genetic variance of progenies of each pair when determining the number of progenies per each pair for each generation
nNextPopVec[numeric] Number of progenies for the next generation for each generation
nameMethod[character] Method for naming individuals
nCores[numeric] Number of cores used for simulations of breedinhh scheme
nCoresPerOptimization[numeric] Number of cores used for simulations of breeding scheme when optimizing hyperparameters
overWriteRes[logical] Overwrite simulation results when the targeted results already exists
showProgress[logical] Show progress bar or not
returnMethod[character] Which type of results will be returned (saved) in the object
saveAllResAt[character] If NULL, we won't save the simulation results. Else, we will save bsInfo and breederInfo for each iteration in the directory defined by 'saveAllResAt' path.
evaluateGVMethod[character] Which type of GV (true GV or estimated GV) will be used for evaluation of the simulation results.
nTopEval[numeric] Number of individuals to be evaluated when evaluating population max or population min
traitNoEval[numeric] Trait No of your interest for evaluation of the method
hEval[numeric] Hyperparameter which determines which BV is emphasized when evaluating simulation results of each method
summaryAllResAt[character] If NULL, we will summary the simulation results in 'self$trueGVMatList' and 'self$estimatedGVMatList'. Else, we will summary the simulation results in 'summaryAllResAt' path.
verbose[logical] Display info (optional)
lociEffectsInit[matrix] Marker and QTL effects used for crossInfo object for initial population
hLens[numeric] Length of hyperparameter vector
hStart[numeric] Initial values of hyper parameters
solnInit[list] List of solution of StoSOO for initial state
hVecOptsList[list] List of optimized hyperparameters in each generation
optimalHyperParamMatsList[list] List of optimized hyper parameters given finite numbers of budget for optimization
simBsRes[list] Simulation results of each method
trueGVMatInit[matrix] A true GV matrix for initial population
trueGVMatList[list] A list of true GV matrix of simulation results
trueGVSummaryArray[array] An array of summary statistics of true GVs for each population & each iteration
estimatedGVMatInit[matrix] An estimated GV matrix for initial population
estimatedGVMatList[list] A list of estimated GV matrix of simulation results
estimatedGVSummaryArray[array] An array of summary statistics of estimated GVs for each population & each iteration
Methods
Public methods
Method new()
Create a new simBsOpt object.
Usage
simBsOpt$new( simBsName = "Undefined", bsInfoInit, breederInfoInit = NULL, lociEffMethod = NULL, trainingPopType = NULL, trainingPopInit = NULL, trainingIndNamesInit = NULL, methodMLRInit = NULL, multiTraitInit = FALSE, nIterSimulation = NULL, nGenerationProceed = NULL, nGenerationProceedSimulation = NULL, setGoalAsFinalGeneration = TRUE, performOptimization = NULL, useFirstOptimizedValue = NULL, sameAcrossGeneration = TRUE, hMin = NULL, hMax = NULL, nTotalIterForOneOptimization = NULL, nIterSimulationPerEvaluation = NULL, nIterOptimization = NULL, nMaxEvalPerNode = NULL, maxDepth = NULL, nChildrenPerExpansion = NULL, confidenceParam = NULL, returnOptimalNodes = NULL, saveTreeNameBase = NULL, whenToSaveTrees = NA, nTopEvalForOpt = NULL, rewardWeightVec = NULL, digitsEval = NULL, nRefreshMemoryEvery = NULL, updateBreederInfo = TRUE, phenotypingInds = FALSE, nRepForPhenoInit = NULL, nRepForPheno = NULL, updateModels = FALSE, updateBreederInfoSimulation = TRUE, phenotypingIndsSimulation = FALSE, nRepForPhenoSimulation = NULL, updateModelsSimulation = FALSE, methodMLR = NULL, multiTrait = FALSE, nSelectionWaysVec = NULL, selectionMethodList = NULL, traitNoSelList = NULL, blockSplitMethod = NULL, nMrkInBlock = NULL, minimumSegmentLength = NULL, nSelInitOPVList = NULL, nIterOPV = NULL, nProgeniesEMBVVec = NULL, nIterEMBV = NULL, nCoresEMBV = NULL, clusteringForSelList = FALSE, nClusterList = NULL, nTopClusterList = NULL, nTopEachList = NULL, nSelList = NULL, multiTraitsEvalMethodList = NULL, hSelList = NULL, matingMethodVec = NULL, allocateMethodVec = NULL, weightedAllocationMethodList = NULL, traitNoRAList = NULL, includeGVPVec = FALSE, nNextPopVec = NULL, nameMethod = "pairBase", nCores = NULL, nCoresPerOptimization = NULL, overWriteRes = FALSE, showProgress = TRUE, returnMethod = "all", saveAllResAt = NULL, evaluateGVMethod = "true", nTopEval = NULL, traitNoEval = NULL, hEval = NULL, summaryAllResAt = NULL, verbose = TRUE )
Arguments
simBsName[character] Name of this simulation of breeding schemes
bsInfoInit[bsInfo] breeding scheme info (see:bsInfo)
breederInfoInit[breederInfo] breeder info (see:breederInfo)
lociEffMethod[character] How to compute loci effects (use true QTL effects ("true") or estimated marker effects ("estimated"))
trainingPopType[character] Training population consists of all populations created in the breeding scheme for 'trainingPopType = "all"', or consists of the latest population in breeding scheme for 'trainingPopType = "latest"'.
trainingPopInit[character / numeric] Training population names or No.s (not generations!!) for initial population
trainingIndNamesInit[character] Names of training individuals for initia; population
methodMLRInit[character] Methods for estimating marker effects for initial population. The following methods are offered:
"Ridge", "LASSO", "ElasticNet", "RR-BLUP", "GBLUP", "BayesA" (uni-trait), "BayesB" (uni-trait), "BayesC" (uni-trait), "BRR", "BL" (uni-trait), "SpikeSlab" (multi-trait)
multiTraitInit[logical] Use multiple-trait model for estimation of marker effects or not for initial population
nIterSimulation[numeric] Number of iterations for this simulation setting
nGenerationProceed[numeric] Number of generations to be proceeded
nGenerationProceedSimulation[numeric] Number of generations to be proceeded in simulation for optimizing policy
setGoalAsFinalGeneration[logical] Set goal for simulation of breeding scheme as the real final generation
performOptimization[logical] Perform optimization of policy in each generation or not
useFirstOptimizedValue[logical] Perform optimization of policy only for the initial population and use the optimized hyperprameters permanently.
sameAcrossGeneration[logical] Use same hyper prameter across generations or not
hMin[numeric] Lower bound of hyper parameters
hMax[numeric] Upper bound of hyper parameters
nTotalIterForOneOptimization[numeric] Number of total iterations that can be assigned for one optimization process
nIterSimulationPerEvaluation[numeric] Number of simulations per one evaluation of the hyperparameter set of your interest
nIterOptimization[numeric] Number of iterations required for one optimization
nMaxEvalPerNode[numeric] Number of maximum evaluation per node when optimizing hyper parameter by StoSOO
maxDepth[numeric] Maximum depth of tree when optimizing hyper parameter by StoSOO
nChildrenPerExpansion[numeric] Number of children per one expansion of nodes when optimizing hyper parameter by StoSOO
confidenceParam[numeric] Confidence parameter of StoSOO, this parameter determines the width of the estimates of rewards
returnOptimalNodes[numeric] When (how many iterations) to return (or save) the optimal nodes when optimizing hyper parameter by StoSOO
saveTreeNameBase[character] Base name of the tree to be saved
whenToSaveTrees[numeric] When (how many iterations) to save the tree in StoSOO
nTopEvalForOpt[numeric] Number of individuals to be evaluated when evaluating population max or population min for optimization of hyperparameters
rewardWeightVec[numeric] When returning reward function, 'rewardWeightVec' will be multiplied by estimated GVs for each generation to evaluate the method. If you want to apply discounted method, you can achieve by 'rewardWeightVec = sapply(1:nGenerationProceedSimulation, function(genProceedNo) gamma ^ (genProceedNo - 1))' where 'gamma' is discounted rate. Or if you want to evaluate just the final generation, you can achieve by 'rewardWeightVec = c(rep(0, nGenerationProceedSimulation - 1), 1)'.
digitsEval[numeric] When you evaluate each hyperparameter, you can round the average of evaluates ('eval') with 'round(eval, digitsEval)'.
nRefreshMemoryEvery[numeric] Every 'nRefreshMemoryEvery' iterations, we refresh memory used for simulations by 'gc(reset = TRUE)'.
updateBreederInfo[logical] Update breederInfo or not for each generation
phenotypingInds[logical] Phenotyping individuals or not for each generation
nRepForPhenoInit[numeric] Number of replications to be phenotyped for initial population
nRepForPheno[numeric] Number of replications to be phenotyped for each generation
updateModels[logical] Whether or not updating the model for each generation (If 'phenotypingInds = FALSE' for generation of your interest, 'updateModels' will be automatically 'FALSE' for that generation.)
updateBreederInfoSimulation[logical] Update breederInfo or not for each generation in simulations for optimization
phenotypingIndsSimulation[logical] Phenotyping individuals or not for each generation in simulations for optimization
nRepForPhenoSimulation[numeric] Number of replications to be phenotyped for each generation in simulations for optimization
updateModelsSimulation[logical] Whether or not updating the model for each generation in simulations for optimization (If 'phenotypingInds = FALSE' for generation of your interest, 'updateModels' will be automatically 'FALSE' for that generation.)
methodMLR[character] Methods for estimating marker effects. The following methods are offered:
"Ridge", "LASSO", "ElasticNet", "RR-BLUP", "GBLUP", "BayesA" (uni-trait), "BayesB" (uni-trait), "BayesC" (uni-trait), "BRR", "BL" (uni-trait), "SpikeSlab" (multi-trait)
multiTrait[logical] Use multiple-trait model for estimation of marker effects or not
nSelectionWaysVec[numeric] Number of selection ways for each generation
selectionMethodList[list] (list of) Selection method for each generation
traitNoSelList[list] (list of list of) Number of trait No of your interest for selection for each generation
blockSplitMethod[character] How to determine the markers belonging to each block when computing OHV. You can define the number of markers in each block ('nMrkInBlock') or the minimum length of each segment ('minimumSegmentLength').
nMrkInBlock[numeric] Number of markers in each block. This will be used for the computation of OHV.
minimumSegmentLength[numeric] Minimum length of each segment [cM]. This will be used for the computation of OHV.
nSelInitOPVList[list] (list of) Number of selected candiates for first screening before selecting parent candidates by OPV for each generation
nIterOPV[numeric] Number of iterations for computation of OPV
nProgeniesEMBVVec[numeric] Number of progenies of double haploids produced when computing EMBV for each generation
nIterEMBV[numeric] Number of iterations to estimate EMBV
nCoresEMBV[numeric] Number of cores used for EMBV estimation
clusteringForSelList[list] (list of) Apply clustering results of marker genotype to selection or not for each generation
nClusterList[list] (list of) Number of clusters for each generation
nTopClusterList[list] (list of) Number of top clusters used for selection for each generation
nTopEachList[list] (list of) Number of selected individuals in each cluster for each generation
nSelList[list] (list of) Number of selection candidates for each generation
multiTraitsEvalMethodList[list] (list of) When evaluating multiple traits, you can choose how to evaluate these traits simultaneously. One method is to take a weighted sum of these traits, and the other is to compute a product of these traits (adjusted to be positive values in advance.)
hSelList[list] (list of) Hyperparameter which determines which trait is weighted when selecting parent candidates for multiple traits for each generation
matingMethodVec[character] Mating method for each generation
allocateMethodVec[character] Allocation method for each generation
weightedAllocationMethodList[list] (list of) Which selection index will be used for weighted resource allocation for each generation
traitNoRAList[list] (list of) Trait No of your interest for resource allocation for each generation
includeGVPVec[logical] Whether or not to consider genetic variance of progenies of each pair when determining the number of progenies per each pair for each generation
nNextPopVec[numeric] Number of progenies for the next generation for each generation
nameMethod[character] Method for naming individuals
nCores[numeric] Number of cores used for simulations of breeding scheme
nCoresPerOptimization[numeric] Number of cores used for simulations of breeding scheme when optimizing hyperparameters
overWriteRes[logical] Overwrite simulation results when the targeted results already exists
showProgress[logical] Show progress bar or not
returnMethod[character] Which type of results will be returned (saved) in the object
saveAllResAt[character] If NULL, we won't save the simulation results. Else, we will save bsInfo and breederInfo for each iteration in the directory defined by 'saveAllResAt' path.
evaluateGVMethod[character] Which type of GV (true GV or estimated GV) will be used for evaluation of the simulation results
nTopEval[numeric] Number of individuals to be evaluated when evaluating population max or population min#'
traitNoEval[numeric] Trait No of your interest for evaluation of the method
hEval[numeric] Hyperparameter which determines which BV is emphasized when evaluating simulation results of each method
summaryAllResAt[character] If NULL, we will summary the simulation results in 'self$trueGVMatList' and 'self$estimatedGVMatList'. Else, we will summary the simulation results in 'summaryAllResAt' path.
verbose[logical] Display info (optional)
Returns
A new 'simBs' object.
Examples
### create simulation information
mySimInfo <- simInfo$new(simName = "Simulation Example",
simGeno = TRUE,
simPheno = TRUE,
#' nSimGeno = 1,
#' nSimPheno = 3,
#' nCoreMax = 4,
#' nCorePerGeno = 1,
#' nCorePerPheno = 3,
saveDataFileName = NULL)
### create specie information
mySpec <- specie$new(nChr = 3,
lChr = c(100, 150, 200),
specName = "Example 1",
ploidy = 2,
mutRate = 10^-8,
recombRate = 10^-6,
chrNames = c("C1", "C2", "C3"),
nLoci = 100,
recombRateOneVal = FALSE,
effPopSize = 100,
simInfo = mySimInfo,
verbose = TRUE)
### create lociInfo object
myLoci <- lociInfo$new(genoMap = NULL, specie = mySpec)
plot(myLoci, alpha = 0.1)
### create traitInfo object
myTrait <- traitInfo$new(lociInfo = myLoci,
nMarkers = 80,
nTraits = 3,
nQTLs = c(4, 8, 3),
actionTypeEpiSimple = TRUE,
qtlOverlap = TRUE,
nOverlap = c(2, 3, 0),
effCor = 0.1,
propDomi = 0.2,
interactionMean = c(4, 1, 2))
myTrait$plot(alphaMarker = 0.1)
### create bsInfo object
myBsInfo <- bsInfo$new(simInfo = mySimInfo,
specie = mySpec,
lociInfo = myLoci,
traitInfo = myTrait,
geno = NULL,
haplo = NULL,
founderIsInitPop = TRUE,
popNameBase = "Population",
initIndNames = NULL,
verbose = TRUE)
### create cross information object
for (i in 1:10) {
myCrossInfo <- crossInfo$new(parentPopulation = myBsInfo$populations[[myBsInfo$generation]],
method = "randomMate",
nNextPop = 100)
myBsInfo$nextGeneration(crossInfo = myCrossInfo)
}
geno <- myBsInfo$overGeneration()$genoMat
myBsInfo$print()
myBsInfo$plot(plotTarget = "trueAGV",
targetTrait = 1:3,
targetPopulation = 1:11,
plotType = "jitter")
Method computeLociEffInit()
estimate/extract marker/QTL effects information
Usage
simBsOpt$computeLociEffInit()
Method startSimulation()
start simulation of breeding scheme
Usage
simBsOpt$startSimulation()
Method summaryResults()
start simulation of breeding scheme
Usage
simBsOpt$summaryResults()
Method print()
Display information about the object
Usage
simBsOpt$print()
Method plot()
Draw figures for visualization of simulation results for summary statistics
Usage
simBsOpt$plot( targetTrait = 1, targetPopulation = NULL, plotType = "box", plotTargetDensity = "max", plotGVMethod = "true", adjust = 1e-05 )
Arguments
targetTrait[numeric] Target trait. character is OK, but numeric vector corresponding to target traits is preferred. It should be a vector with length 1.
targetPopulation[numeric] Target populations. character is OK, but numeric vector corresponding to target traits is preferred.
plotType[character] We offer "box", "violin", "lines", "density" to draw figures for simulation results.
plotTargetDensity[character] When you choose "density" for 'plotType', you should select which summary statistics will be plotted. It should be a vector with length 1.
plotGVMethod[character] Which type of GV (true GV or estimated GV) will be used for plotting the simulation results
adjust[numeric] the bandwidth used is actually adjust*bw. This makes it easy to specify values like ‘half the default’ bandwidth. (see: 'adjust' in density)
Method clone()
The objects of this class are cloneable with this method.
Usage
simBsOpt$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.
Examples
## ------------------------------------------------
## Method `simBsOpt$new`
## ------------------------------------------------
### create simulation information
mySimInfo <- simInfo$new(simName = "Simulation Example",
simGeno = TRUE,
simPheno = TRUE,
#' nSimGeno = 1,
#' nSimPheno = 3,
#' nCoreMax = 4,
#' nCorePerGeno = 1,
#' nCorePerPheno = 3,
saveDataFileName = NULL)
### create specie information
mySpec <- specie$new(nChr = 3,
lChr = c(100, 150, 200),
specName = "Example 1",
ploidy = 2,
mutRate = 10^-8,
recombRate = 10^-6,
chrNames = c("C1", "C2", "C3"),
nLoci = 100,
recombRateOneVal = FALSE,
effPopSize = 100,
simInfo = mySimInfo,
verbose = TRUE)
### create lociInfo object
myLoci <- lociInfo$new(genoMap = NULL, specie = mySpec)
plot(myLoci, alpha = 0.1)
### create traitInfo object
myTrait <- traitInfo$new(lociInfo = myLoci,
nMarkers = 80,
nTraits = 3,
nQTLs = c(4, 8, 3),
actionTypeEpiSimple = TRUE,
qtlOverlap = TRUE,
nOverlap = c(2, 3, 0),
effCor = 0.1,
propDomi = 0.2,
interactionMean = c(4, 1, 2))
myTrait$plot(alphaMarker = 0.1)
### create bsInfo object
myBsInfo <- bsInfo$new(simInfo = mySimInfo,
specie = mySpec,
lociInfo = myLoci,
traitInfo = myTrait,
geno = NULL,
haplo = NULL,
founderIsInitPop = TRUE,
popNameBase = "Population",
initIndNames = NULL,
verbose = TRUE)
### create cross information object
for (i in 1:10) {
myCrossInfo <- crossInfo$new(parentPopulation = myBsInfo$populations[[myBsInfo$generation]],
method = "randomMate",
nNextPop = 100)
myBsInfo$nextGeneration(crossInfo = myCrossInfo)
}
geno <- myBsInfo$overGeneration()$genoMat
myBsInfo$print()
myBsInfo$plot(plotTarget = "trueAGV",
targetTrait = 1:3,
targetPopulation = 1:11,
plotType = "jitter")
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.