Package 'PhenStat'

Description

Method analysedDataset returns subset of the original dataset that was analysed

Value

This function returns the analysed dataset.

Methods

signature(obj = "PhenTestResult")

Examples

# Mixed Model framework
    file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")

    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    head(PhenStat:::analysedDataset(obj = result))

Description

Method analysedDatasetPhenList returns subset of the original dataset that will be analysed

Value

The function returns the analysed dataset.

Methods

signature(obj = "PhenList")

Description

Method analysedSubset returns name of the subset that was analysed by RR or FE frameworks

Value

The function returns name of the subset that was analysed by RR or FE frameworks.

Methods

signature(obj = "htestPhenStat")

Description

Method analysisResults returns analysis results

Value

The function returns the analysis result.

Methods

signature(obj = "PhenTestResult")

Examples

# Mixed Model framework
    file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
                                testGenotype="Sparc/Sparc")

    result <- PhenStat:::testDataset(test,
                                     depVariable="Lean.Mass")
    r = PhenStat:::analysisResults(obj = result)

Description

Method batchIn returns TRUE/FALSE values depending on the batch column presence/absence in the dataset

Value

TRUE/FALSE

Methods

signature(obj = "PhenList")

signature(obj = "PhenTestResult")

Description

Graph function for the Mixed Model framework's results. Creates a box plot with residue versus batch split by genotype.

Usage

boxplotResidualBatch(phenTestResult,outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList and PhenTestResult

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    PhenStat:::boxplotResidualBatch(result)

Description

Graph function for the phenotypic dataset. Creates a box plot split by sex and genotype.

Usage

boxplotSexGenotype(phenList,
                     depVariable=NULL,
                     graphingName=NULL,
                     outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
  test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
                              testGenotype="Sparc/Sparc")
  # box plot for dataset with two sexes: males and females
  PhenStat:::boxplotSexGenotype(test,
                                depVariable="Bone.Mineral.Content",
                                graphingName="BMC")

  file <- system.file("extdata", "test4.csv", package="PhenStat")
  test_1sex <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
                                   testGenotype="Mysm1/+")
  # box plot for females only dataset
  PhenStat:::boxplotSexGenotype(test_1sex,
                                depVariable="Lean.Mass",
                                graphingName="Lean Mass (g)")

Description

NB! Function is deprecated, please use scatterplotSexGenotypeBatch function instead. Graph function for the phenotypic dataset. Creates a box plot split by sex, genotype and batch.

Note: the batches are not ordered with time but allow assessment of how the treatment groups lie relative to the normal control variation.

Usage

boxplotSexGenotypeBatch(phenList,
							depVariable=NULL,
							graphingName=NULL,
							outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    # box plot for dataset with two sexes: males and females
    PhenStat:::boxplotSexGenotypeBatch(test,
            depVariable="Bone.Mineral.Content",
            graphingName="BMC" )

    file <- system.file("extdata", "test4.csv", package="PhenStat")
    test_1sex <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mysm1/+")
    # box plot for females only dataset
    PhenStat:::boxplotSexGenotypeBatch(test_1sex,depVariable="Lean.Mass",
            graphingName="Lean Mass")

Description

Graph function for the phenotypic dataset. Creates a box plot split by sex and genotype after accounting for batch.

Usage

boxplotSexGenotypeBatchAdjusted(
		phenList,
		depVariable=NULL,
		graphingName=NULL,
		outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    # box plot for dataset with two sexes: males and females
    PhenStat:::boxplotSexGenotypeBatchAdjusted(test,
            depVariable="Bone.Mineral.Content",
            graphingName="BMC" )

Description

Graph function for the phenotypic dataset. Creates a box plot split by sex and genotype. Is based on PhenTestResult object.

Usage

boxplotSexGenotypeResult(phenTestResult,
							graphingName=NULL,
							outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    # box plot for dataset with two sexes: males and females
    PhenStat:::boxplotSexGenotypeResult(result,
            graphingName="BMC")

    file <- system.file("extdata", "test4.csv", package="PhenStat")
    test_1sex <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mysm1/+")
    result_1sex <- PhenStat:::testDataset(test_1sex,
            depVariable="Lean.Mass")
    # box plot for females only dataset
    PhenStat:::boxplotSexGenotypeResult(result_1sex,
            graphingName="Lean Mass (g)")

Description

Graph function for the phenotypic dataset. Creates a box plot split by sex and genotype after accounting for batch and weight.

Usage

boxplotSexGenotypeWeightBatchAdjusted(
		phenList,
		depVariable=NULL,
		graphingName=NULL,
		outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    # box plot for dataset with two sexes: males and females
    PhenStat:::boxplotSexGenotypeWeightBatchAdjusted(test,
            depVariable="Bone.Mineral.Content",
            graphingName="BMC" )

Description

Graph function for the Fisher Exact Test framework's results. Creates stacked bar plot(s) to compare proportions seen in a categorical variable between different genotypes. Graphs are created for all data and also for each sex individually.

Usage

categoricalBarplot(phenTestResult,
					   outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test_categorical.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Aff3/Aff3")
    result <- PhenStat:::testDataset(test,
            depVariable="Thoracic.Processes",
            method="FE")
    PhenStat:::categoricalBarplot(result)

Description

Checks dataset for the minimum required info:

1. Column names should present

2. Genotype column should present

3. Sex column should present

4. Two data points for each Genotype/Sex combination

5. Number of Genotype levels should be 2

6. Number of Sex levels should be 1 or 2

7. Sex levels have to be "Female" and/or "Male"

8. Records with reference genotype should be in the dataset

9. Records with test genotype should be in the dataset

Perform the following additional checks:

- presence of Weight column,

- presence of Batch column (Assay.Date).

Warning given in case of missed data indicating that you can only fit a glm or to use MM equation "withoutWeight".

Function checkDataset is called from PhenList function.

Usage

checkDataset(dataset, 
				testGenotype, 
				refGenotype="+/+",
				outputMessages=TRUE,
				dataset.clean=TRUE)

Arguments

Value

Returns an instance of the PhenList class.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

# "checkDataset" is called internally from "PhenList" function
    file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")

Description

Returns a classification tag to assign a sexual dimorphism assessment of the phenotypic change.

Usage

classificationTag(phenTestResult, 
					userMode = "summaryOutput",
					phenotypeThreshold = 0.01,
					outputMessages=TRUE)

Arguments

Value

Returns a classification tag to assign a sexual dimorphism assessment of the phenotypic change.

If you are working interactively with the data, the argument "userMode" set to the value "summaryOutput" will use the "phenotypeThreshold" argument's value to assess statistical significance of the genotype effect and if significant then assign a sexual dimorphic classification. Alternatively, if the "userMode" set to the value "vectorMode", a sexual dimorphic classification will be returned with the MM framework where later you can globally assess whether the variable had a significant genotype effect. With the FE framework and the vectorMode, a NA is returned as the type of the effect cannot be assessed without assessing the statistical significance of the genotype effect.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    PhenStat:::classificationTag(result, 
            userMode="summaryOutput", 
            phenotypeThreshold=0.001)
    PhenStat:::classificationTag(result, 
            userMode="vectorOutput")

Description

Method comparison returns name of the comparison that was performed by RR framework: "High vs Normal/Low" or "Low vs Normal/High"

Value

returns name of the comparison that was performed by RR framework.

Methods

signature(obj = "htestPhenStat")

Description

Method dataset returns data frame stored in the object

Value

returns data frame stored in the object

Methods

signature(obj = "PhenList")

Description

This is an internal function run within LR framework. It completes the final stage of the LR framework, which builds the final model and estimates effects. As an internal function, it doesn't include extensive error testing of inputs. Please use cautiously if calling directly.

Works with PhenTestResult object created by startLRModel function. The creation of LR final model is based on the significance of different fixed effects, depVariable and equation values stored in PhenTestResult object.

Usage

finalLRModel(phenTestResult, 
				outputMessages = TRUE)

Arguments

Value

Returns results stored in instance of the PhenTestResult class

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult and testDataset

Examples

file <- system.file("extdata", "testLR.csv", package="PhenStat")
    testLR <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mock")
    testLR2 <- PhenStat:::LRDataset(testLR,"V2")
    
    # when "testDataset" function's argument "callAll" is set to FALSE 
    # only "startLRModel" function is called - the first step of LR framework
    resultLR <- PhenStat:::testDataset(testLR2,
            depVariable="V2",
            callAll=FALSE,
            method="LR")
    # print out formula that has been created
    PhenStat:::analysisResults(resultLR)$model.formula.genotype
    # print out batch effect's significance 
    PhenStat:::analysisResults(resultLR)$model.effect.batch
    resultLR <- PhenStat:::finalLRModel(resultLR)

Description

This is an internal function run within MM framework. It completes the final stage of the MM framework, which builds the final model and estimates effects. As an internal function, it doesn't include extensive error testing of inputs. Please use cautiously if calling directly.

Works with PhenTestResult object created by startModel function. The creation of MM final model is based on the significance of different fixed effects, depVariable and equation values stored in PhenTestResult object.

Usage

finalModel(phenTestResult,
				outputMessages = TRUE,
				modelWeight = NULL)

Arguments

Value

Returns results stored in instance of the PhenTestResult class

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult and testDataset

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    # when "testDataset" function's argument "callAll" is set to FALSE
    # only "startModel" function is called - the first step of MM framework
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass",
            equation="withoutWeight",
            callAll=FALSE)
    # print out formula that has been created
    # result$model.formula.genotype
    # print out batch effect's significance
    # result$model.effect.batch
    # change the model
    # result <- PhenStat:::testDataset(test,
    #        depVariable="Lean.Mass",
    #        equation="withWeight",
    #        callAll=FALSE)
    # print out new formula
    #result$model.formula.genotype
    # run the final model fitting when statisfied with the model
    result <- PhenStat:::finalModel(result)

Description

This is an internal function run within TF framework. It completes the final stage of the TF framework, which builds the final model and estimates effects. As an internal function, it doesn't include extensive error testing of inputs. Please use cautiously if calling directly.

Works with PhenTestResult object created by startTFModel function. The creation of TF final model is based on the significance of different fixed effects, depVariable and equation values stored in PhenTestResult object.

Usage

finalTFModel(phenTestResult, 
				outputMessages = TRUE)

Arguments

Value

Returns results stored in instance of the PhenTestResult class

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult and testDataset

Examples

file <- system.file("extdata", "test7_TFE.csv", package="PhenStat")
    test <- PhenList(dataset=read.csv(file,na.strings = '-'),
                     testGenotype="het",
                     refGenotype = "WT",
                     dataset.colname.sex="sex",
                     dataset.colname.genotype="Genotype",
                     dataset.values.female="f",
                     dataset.values.male= "m",
                     dataset.colname.weight="body.weight",
                     dataset.colname.batch="Date_of_procedure_start")

    test_TF <- PhenStat:::TFDataset(test,depVariable="Cholesterol")
    
    # when "testDataset" function's argument "callAll" is set to FALSE 
    # only "startTFModel" function is called - the first step of TFE framework
    result  <- PhenStat:::testDataset(test_TF,
                           depVariable="Cholesterol",
                           callAll=FALSE,
                           method="TF")
    # print out formula that has been created
    PhenStat:::analysisResults(result)$model.formula.genotype
    # print out batch effect's significance 
    PhenStat:::analysisResults(result)$model.effect.batch
    result <- PhenStat:::finalTFModel(result)

Description

The main function of the Fisher Exact Test framework. Creates n times 2 matrices with record counts, where n rows represent dependent variable levels and two columns represent genotype levels. Performs Fisher Exact Tests on calculated count matrices.

Three matrices can be created and three tests can be potentially perform depending on the dataset:

- all records together regardless the sex values - combined dataset,

- records where sex value is "Male" (if such exists) - males only dataset,

- records where sex value is "Female" (if such exists) - females only dataset.

Together with count matrices creates percentage matrices, calculates effect sizes and statistics for count matrices. Performs Fisher Exact Tests.

The results (matrices, statistics and Fisher Exact Test outputs) are stored in PhenTestResult object.

The funciton is called from "testDataset" function when "method" argument is set to "FE" meaning "Fisher Exact Test".

Usage

FisherExactTest(phenList, 
					depVariable, 
					outputMessages=TRUE)

Arguments

Value

Returns results stored in instance of the PhenTestResult class

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

file <- system.file("extdata", "test_categorical.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Aff3/Aff3")
    # "FisherExactTest" function is called from "testDataset" function.
    result <- PhenStat:::testDataset(test,depVariable="Thoracic.Processes",method="FE")  
    # Fisher Exact Test results can be printed out using function "summaryOutput"
    summary(result)

Description

Function generates graphs for the Mixed Model and Time as Fixed Effect frameworks' results and stores generated graphs in the defined directory.

Usage

generateGraphs(phenTestResult, dir,
            graphingName=NULL, type="Xlib")

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Description

Method getColumn returns column of interest

Value

This function returns column of interest

Methods

signature(obj = "PhenList")

Description

Method getColumnBatchAdjusted returns column of interest adjusted for batch

Value

This function returns column of interest

Methods

signature(obj = "PhenList")

Description

Method getColumnView returns RR or FE frameworks results in a column format

Value

This function returns RR or FE frameworks results in a column format

Methods

signature(obj = "htestPhenStat")

Description

Method getColumnWeightBatchAdjusted returns column of interest adjusted for weight and batch

Value

This function returns column of interest

Methods

signature(obj = "PhenList")

Description

Method getCountMatrices returns count matrices for FE and RR frameworks

Value

This function returns count matrices for FE and RR frameworks

Methods

signature(obj = "PhenTestResult")

Description

Method getDataset returns data frame stored in the object

Value

This function returns data frame stored in the object.

Methods

signature(obj = "PhenList")

Description

Method getGenotypeEffect returns genotype effect and standard error for the linear regression frameworks

Value

This function returns genotype effect and standard error for the linear regression frameworks

Methods

signature(obj = "PhenTestResult")

Description

Method getMatrix returns RR or FE frameworks results in a matrix format

Value

This function returns RR or FE frameworks results in a matrix format

Methods

signature(obj = "htestPhenStat")

Description

Method getPercentageMatrix returns percentage matrix calculated out of counts

Value

This function returns percentage matrix calculated out of counts.

Methods

signature(obj = "htestPhenStat")

Description

Method getStat returns simple statistics about variables in the dataset

Value

This function returns simple statistics about variables in the dataset

Methods

signature(obj = "PhenList")

Description

Method getVariable returns dependent variable name

Value

This function returns dependent variable name

Methods

signature(obj = "PhenTestResult")

Description

Method getVariables returns names of variables in the dataset

Value

This function returns names of variables in the dataset

Methods

signature(obj = "PhenList")

Description

Method hemiGenotype returns hemi genotype if defined

Value

This function returns hemi genotype if defined

Methods

signature(obj = "PhenList")

signature(obj = "PhenTestResult")

Description

Wrapper for the model fitting results. Returns model fitting and testing results in a JSON format. Assumes that modelling results are stored in the PhenTestResult object (output from function testDataset).

Usage

JSONOutput(phenTestResult,
				phenotypeThreshold = 0.01)

Arguments

Value

Returns a vector with model fitting results in JSON format: model output summary and other values

Author(s)

Natalja Kurbatova

See Also

PhenTestResult

Examples

## Not run:
    # Mixed Model framework
    file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    PhenStat:::JSONOutput(result)

    # Fishe Exact Test framework
    file <- system.file("extdata", "test_categorical.csv", package="PhenStat")
    test2 <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Aff3/Aff3")
    result2 <- PhenStat:::testDataset(test2,
            depVariable="Thoracic.Processes",
            method="FE")
    PhenStat:::JSONOutput(result2)
	## End(Not run)

Description

Prepares dataset for the LR framework - maps values of dependent variable to 0/1, where 1 is modeled.

Usage

LRDataset(
  phenList = NULL,
  depVariable = NULL,
  abnormalValues =
    c("abnormal", "Abnormal", "TRUE", "deviant"),
  outputMessages = TRUE
)

Arguments

Value

Returns dataset suitable for LR framework, where provided abonormal values are mapped to 1 and all other values of dependent variable are assumed to be normal and are mapped to 0.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

Examples

file <- system.file("extdata", "testLR.csv", package="PhenStat")
    testLR <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mock")
    testLR2 <- PhenStat:::LRDataset(testLR,"V2")

Description

Method matrixCount returns matrix of counts that was created by RR or FE framework.

Value

This function returns matrix of counts that was created by RR or FE framework.

Methods

signature(obj = "htestPhenStat")

Description

Method method returns method abbriviation, for example: "RR", "MM".

Value

This function returns returns method abbriviation.

Methods

signature(obj = "PhenTestResult")

Description

Method methodText returns method's name, for example: "Reference Range Plus", "Linear Mixed Model".

Value

This function returns returns method's name.

Methods

signature(obj = "PhenTestResult")

Description

Method multipleBatches returns TRUE if there are multiple batches in the dataset, FALSE otherwise

Value

This function returns TRUE if there are multiple batches in the dataset.

Methods

signature(obj = "PhenList")

Description

Method noSexes returns number of sex levels: 1/2

Value

This function returns number of sex levels.

Methods

signature(obj = "PhenList")

signature(obj = "PhenTestResult")

Description

Method parameters returns parameters specific for apllied method/dataset combination

Value

This function returns parameters specific for apllied method/dataset combination.

Methods

signature(obj = "PhenTestResult")

Description

Parsers model output summary and returns it in readable named vector format

Usage

parserOutputSummary(linearRegressionOutput)

Arguments

Value

Returns a named vector with linear regression model output summary results

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    linearRegressionOutput <- PhenStat:::analysisResults(result)
    PhenStat:::parserOutputSummary(linearRegressionOutput)

Description

Parsers model output summary and returns it in readable named vector format

Usage

parserOutputSummaryLR(linearRegressionOutput)

Arguments

Value

Returns a named vector with logistic regression model output summary results

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

file <- system.file("extdata", "testLR.csv", package="PhenStat")
    testLR <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mock")
    testLR2 <- PhenStat:::LRDataset(testLR,"V2")
    resultLR <- PhenStat:::testDataset(testLR2,
            depVariable="V2",         
            method="LR")
    linearRegressionOutput <- PhenStat:::analysisResults(resultLR)
    PhenStat:::parserOutputSummaryLR(linearRegressionOutput)

Description

Parsers model output summary from TF framework and returns it in readable named vector format

Usage

parserOutputTFSummary(linearRegressionOutput)

Arguments

Value

Returns a named vector with model output summary results

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

file <- system.file("extdata", "test6_RR.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Oxr1/Oxr1")
    result <- PhenStat:::testDataset(test,
            depVariable="Ca",
            method="TF",
            dataPointsThreshold=2)
    linearRegressionOutput <- PhenStat:::analysisResults(result)
    PhenStat:::parserOutputTFSummary(linearRegressionOutput)

Description

Function to create data object from the data frame.

In addition to dependent variable column (the variable of interest) mandatory columns that should present in the data frame are "Genotype" and "Sex". The "Assay.Date" column is used to model "Batch" effect if not specified differently. "Weight" column is used to model body weight effect.

Function creates PhenList object, checks data integrity, renames columns when requested, provides simple statistics about dataset.

Returns an instance of the PhenList object created from the data file.

Dataset is cleaned to ensure there is a maximum two sex levels and there are exactly two levels for genotype. Data cleaning can be switched off by defining the argument "dataset.clean" as FALSE.

Usage

PhenList(dataset, testGenotype, refGenotype="+/+", hemiGenotype=NULL,
            outputMessages=TRUE, dataset.clean=TRUE,
            dataset.colname.batch=NULL, dataset.colname.genotype=NULL,
            dataset.colname.sex=NULL, dataset.colname.weight=NULL,
            dataset.values.missingValue=" ", dataset.values.male=NULL,
            dataset.values.female=NULL)

Arguments

Value

Returns an instance of the PhenList class.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,  na.strings = '-'),
            testGenotype="Sparc/Sparc")

    file <- system.file("extdata", "test2.csv", package="PhenStat")
    test2 <- PhenStat:::PhenList(dataset=read.csv(file, na.strings = '-'),
            testGenotype="Arid4a/Arid4a",
            refGenotype="+/+",
            hemiGenotype="Arid4a/+",
            dataset.colname.weight="Weight.Value")

    file <- system.file("extdata", "test3.csv", package="PhenStat")
    test3 <- PhenStat:::PhenList(dataset=read.csv(file, na.strings = '-'),
            dataset.clean=TRUE,
            dataset.values.female=1,
            dataset.values.male=2,
            testGenotype="Mysm1/+")

Description

A list-based S4 class for storing phenotypic data. Helps to support data integrity checks and statistics calculation. The PhenList object can be created by using function PhenList.

Explore PhenList object

PhenList object instance contains the following slots:

1. "datasetPL" where cleaned and checked dataset is stored: getDataset(phenList)

2. "testGenotype" where the genotype level to test is stored: testGenotype(phenList)

3. "refGenotype" where reference genotype value is stored with default value set to "+/+":

refGenotype(phenList)

4. "hemiGenotype" if defined contains the genotype value for hemizygous: hemiGenotype(phenList)

6. "dataset.clean" flag value is stored, see PhenList for more details:

[email protected]

7. "dataset.colname" if defined contains dataset column names that have been renamed:

- [email protected] column name for batch values

- [email protected] column name for genotype values

- [email protected] column name for sex values

- [email protected] column name for weight values

8. "dataset.values" if defined contains dataset values that have been changed during dataset cleaning process:

- [email protected] value used as missing value in the orginal dataset

- [email protected] value used to label "males" in the original dataset

- [email protected] value used to label "females" in the original dataset

Slots

datasetPL:

Object of class "data.frame" ~~ dataset to work with

refGenotype:

Object of class "character" ~~ reference genotype

testGenotype:

Object of class "character" ~~ test genotype

hemiGenotype:

Object of class "character" ~~ hemi genotype

dataset.colname.batch:

Object of class "character" ~~ column name for batch values

dataset.colname.genotype:

Object of class "character" ~~ column name for genotype values

dataset.colname.sex:

Object of class "character" ~~ column name for sex values

dataset.colname.weight:

Object of class "character" ~~ column name for weigth values

dataset.values.missingValue:

Object of class "character" ~~ value used as missing value in the orginal dataset

dataset.values.male:

Object of class "character" ~~ value used to label "males" in the original dataset

dataset.values.female:

Object of class "character" ~~ value used to label "females" in the original dataset

dataset.clean:

Object of class "logical" ~~ flag value is stored

datasetUNF:

Object of class "data.frame" ~~ unfiltered dataset

Methods

getDataset

(accessor): Returns dataset

refGenotype

(accessor): Returns reference genotype

testGenotype

(accessor): Returns test genotype

hemiGenotype

(accessor): Returns hemi genotype if specified

getColumn

Returns specified column if exists

getColumnBatchAdjusted

Returns specified column adjusted for batch effect

getColumnWeightBatchAdjusted

Returns specified column adjusted for batch and weight effects

getStat

Returns simple dataset statistics

getVariables

Returns dataset column names

batchIn

Returns TRUE if the batch is in the dataset, FALSE otherwise

weightIn

Returns TRUE if the weight is in the dataset, FALSE otherwise

multipleBatches

Returns TRUE if the batches are variable in the dataset, FALSE otherwise

noSexes

Returns number of sexes in the dataset

setBatch

Sets the batch column - change the column names

setGenotype

Sets the genotype column - change the column names

setMissingValue

Sets the missing value

setSex

Sets the sex column - change the column names

setWeight

Sets the weight column - change the column names

Author(s)

Natalja Kurbatova

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    class(test)
    
    file <- system.file("extdata", "test2.csv", package="PhenStat")
    test2 <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Arid4a/Arid4a",
            refGenotype="+/+",
            hemiGenotype="Arid4a/+",
            dataset.colname.weight="Weight.Value")
    PhenStat:::getStat(test2)
    PhenStat:::testGenotype(test2)
    PhenStat:::refGenotype(test2)
    PhenStat:::hemiGenotype(test2)

Description

This function takes a Phenlist object and generates a pdf report containing several statistical methods.

Usage

PhenStatReport(
    PhenlistObject           ,
    depVariable = NULL       ,
    other.response = NULL    ,
    update = TRUE            ,
    Gene.Symbol = NULL       ,
    Response.name = NULL     ,
    destination = NULL       ,
    reportTitle = "Extended Statistical Report",
    DataRelease = NULL       ,
    Showsource = FALSE       ,
    open = FALSE             ,
    clean = TRUE             ,
    verbos = FALSE           ,
    ...
  )

Arguments

Details

See PhenStatReport page on the IMPC website

Value

Author(s)

Hamed Haselimashhadi <[email protected]>

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package = "PhenStat")
    test = PhenStat:::PhenList(dataset = read.csv(file,na.strings = '-'),
                               testGenotype = "Sparc/Sparc")
 ## Not run: 
    PhenStatReport(test,
                   depVariable = 'Bone.Area',
                   open = TRUE)
  
## End(Not run)

Description

Creates PhenTestResult object based on test results or model building results. Internal function for the package. Not build for users to directly call.

Usage

PhenTestResult(model.output=NULL, model.dataset=NULL,
            depVariable=NULL, 
            refGenotype=NULL,
            equation="withWeight", 
            method="MM", model.effect.batch=NULL, 
            model.effect.variance=NULL,model.effect.sex=NULL, 
            model.effect.interaction=NULL, model.output.interaction=NULL, 
            model.effect.weight=NULL, numberSexes=NULL,pThreshold=0.05, 
            model.formula.null=NULL,model.formula.genotype=NULL, 
            model.output.genotype.nulltest.pVal=NULL, 
            model.output.quality=NULL,model.output.summary=NULL,
            model.output.averageRefGenotype = NULL,
            model.output.percentageChanges = NULL)

Arguments

Value

Returns an instance of the PhenTestResult class.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    # "testDataset" function calls "PhenTestResult" function internally
    result <- PhenStat:::testDataset(test,
            equation="withoutWeight",
            depVariable="Lean.Mass")
    class(result)

Description

A list-based S4 class for storing of model fitting results and other information like dependent variable, method used to build a model, etc. Helps to support operations needed for model fitting process and modelling/testing results storage. In the package PhenTestResult object is created by function testDataset.

Explore PhenTestResult object

PhenTestResults object instance contains the following slots:

1. "analysedDataset" contains subdataset that was used for the modelling/testing:

analysedDataset(phenTestResult)

2. "depVariable" contains dependent variable that was tested: getVariable(phenTestResult)

3. "refGenotype" contains reference genotype which is usually is wildtype: refGenotype(phenTestResult)

4. "testGenotype" contains test genotype: testGenotype(phenTestResult)

5. "method" contains method name that was used for modelling/testing: method(phenTestResult)

6. "transformationRequired" contains TRUE/FALSE value indicating transofrmation of dependent variable:

transformationText(phenTestResult) or phenTestResult@transformationRequired

7. "lambdaValue" contains lambda value for the Box-Cox transformation:

transformationText(phenTestResult) or phenTestResult$lambdaValue

8. "scaleShift" contains the value of scale shift for Box-Cox transformation:

transformationText(phenTestResult) or phenTestResult@scaleShift

9. "transformationCode" contains the code of transformation. Possible values:

0 - transformation is not applicable (methods "FE", "RR", "LR") or switched off by user 1 - transformation is not needed (1 is within the 95 2 - log transformation 3 - power transformation 4 - transformation is not performed since optimal lambda value is not found (-5 > lambda > 5)

transformationText(phenTestResult) or phenTestResult@transformationCode

10. "parameters" contains the parameters used during method application, e.g. pThreshold for MM and TF:

parameters(phenTestResult)

11. Modelling/testing results are stored in the sections "analysisResults": analysisResults(phenTestResult)

The contents of the analysisResults slot depend on framework that was used.

The results of MM and TF frameworks: 1. Equation used during linear modeling: analysisResults(phenTestResult)$equation

2. Batch effect significance: analysisResults(phenTestResult)$model.effect.batch

3. Variance equality: analysisResults(phenTestResult)$model.effect.variance

4. Weight effect significance: analysisResults(phenTestResult)$model.effect.weight

5. Sex effect significance: analysisResults(phenTestResult)$model.effect.interaction

6. Evidence of sex effect (p-value): analysisResults(phenTestResult)$model.output.interaction

7. Evidence of genotype effect (p-value): analysisResults(phenTestResult)$model.output.genotype.nulltest.pVal

8. Formula for the final genotype model: analysisResults(phenTestResult)$model.formula.genotype

9. Formula for the final null model: analysisResults(phenTestResult)$model.formula.null

10. Model fitting output: analysisResults(phenTestResult)$model.output

11. Model fitting summary: summary(analysisResults(phenTestResult)$model.output)$tTable

The results of FE and RR frameworks are stored as a list of htestPhenStat S4 objects: analysisResults(phenTestResult)[[1]].

Each one htestPhenStat object contains:

1. Output of Fisher Exact Test: pvalue(analysisResults(phenTestResult)[[1]]).

2. Effect size: ES(analysisResults(phenTestResult)[[1]]).

3. Name of the table analysed (all, males, females): analysedSubset(analysisResults(phenTestResult)[[1]]).

4. Comparison, used for RR only (High vs Normal/Low, Low vs High/Normal): comparison(analysisResults(phenTestResult)[[1]]).

5. Matrix of counts: matrixCount(analysisResults(phenTestResult)[[1]]).

Slots

analysedDataset:

Object of class "data.frame" ~~ analysed dataset

transformationRequired:

Object of class "logical" ~~ flag indicating was or not transformation performed

lambdaValue:

Object of class "numeric" ~~ Box-Cox transform lambda value

scaleShift:

Object of class "numeric" ~~ Box-Cox transform scale shift

transformationCode:

Object of class "numeric" ~~ code explaining the transformation output

depVariable:

Object of class "character" ~~ dependent variable

refGenotype:

Object of class "character" ~~ reference genotype

testGenotype:

Object of class "character" ~~ test genotype

method:

Object of class "character" ~~ analysis method

parameters:

Object of class "matrix" ~~ parameters of method

analysisResults:

Object of class "list" ~~ results of the analysis

Methods

getVariable

(accessor): Returns dependent variable

refGenotype

(accessor): Returns reference genotype

testGenotype

(accessor): Returns test genotype

method

(accessor): Returns method used

methodText

(accessor): Returns full name of the method used

parameters

(accessor): Returns parameteres used during the method application

analysisResults

(accessor): Returns analysis results depending on method

analysedDataset

(accessor): Returns analysed dataset

transformationText

(accessor): Returns the info about transformation

batchIn

Returns TRUE if the batch is in the dataset, FALSE otherwise

weightIn

Returns TRUE if the weight is in the dataset, FALSE otherwise

noSexes

Returns number of sexes in the dataset

show

Prints out the PhenTestResult object

getCountMatrices

Returns count matrices if they are present (for methods "FE" and "RR"), NULL otherwise.

Author(s)

Natalja Kurbatova

Examples

# Mixed Model framework
    file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            equation="withoutWeight",
            depVariable="Bone.Area")
    PhenStat:::getVariable(result)
    PhenStat:::method(result)
    # Batch effect is significant
    PhenStat:::analysisResults(result)$model.effect.batch    
    # Variance homogeneosity 
    PhenStat:::analysisResults(result)$model.effect.variance
    # Weight effect is significant
    PhenStat:::analysisResults(result)$model.effect.weight     
    # Sex effect is significant 
    PhenStat:::analysisResults(result)$model.effect.interaction    
    # Sex effect p-value - the result of the test 
    PhenStat:::analysisResults(result)$model.output.interaction
    # Genotype effect p-value    
    PhenStat:::analysisResults(result)$model.output.genotype.nulltest.pVal
    # Final model formula with genotype
    PhenStat:::analysisResults(result)$model.formula.genotype 
    # Final model formula without genotype    
    PhenStat:::analysisResults(result)$model.formula.null 
    # Final model fitting output
    # result$model.output 
    # Final model fitting summary
    # summary(result$model.output)$tTable 
    
    # Fisher Exact Test framework
    file <- system.file("extdata", "test_categorical.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Aff3/Aff3")
    result <- PhenStat:::testDataset(test,
            depVariable="Thoracic.Processes",
            method="FE") 
    PhenStat:::getVariable(result)
    PhenStat:::method(result)       
    for (i in seq_along(analysisResults(result))) {
        val <- PhenStat:::analysisResults(result)[[i]]
        val
    }

Description

Plot method for objects of class "PhenList".

Usage

## S3 method for class 'PhenList'
plot(
    x                    ,
    depVariable = 'Value',
    graphingName = NULL  ,
    outputMessages = TRUE,
    type = NULL,
    ...
  )

Arguments

Value

Empty.

Author(s)

Hamed Haselimashhadi < [email protected] >

See Also

PhenList,

Examples

example(testDataset)

Description

Plot method for objects of class "phenTestResult".

Usage

## S3 method for class 'PhenTestResult'
plot(
    x                    ,
    graphingName = NULL  ,
    outputMessages = TRUE,
    type = NULL,
    ...
  )

Arguments

Value

Empty.

Author(s)

Hamed Haselimashhadi < [email protected] >

See Also

testDataset

Examples

example(testDataset)

Description

Graph function for the Mixed Model framework's results. Creates predicted versus residual values plots split by genotype.

Usage

plotResidualPredicted(phenTestResult,outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList and PhenTestResult

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    PhenStat:::plotResidualPredicted(result)

Description

Method pvalue returns p-value that was calculate by RR or FE framework.

Value

This function returns p-value that was calculate by RR or FE framework.

Methods

signature(obj = "htestPhenStat")

Description

Graph function for the Mixed Model framework's results. Creates a Q-Q plot of residuals for each genotype.

Usage

qqplotGenotype(phenTestResult,outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList and PhenTestResult

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    PhenStat:::qqplotGenotype(result)

Description

Graph function for the Mixed Model framework's results. Creates a Q-Q plot of blups (best linear unbiased predictions).

Usage

qqplotRandomEffects(phenTestResult,outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList and PhenTestResult

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    PhenStat:::qqplotRandomEffects(result)

Description

Graph function for the Mixed Model framework's results. Creates a Q-Q plot of rotated residuals.

Usage

qqplotRotatedResiduals(phenTestResult,outputMessages=TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

Houseman, E. A., Ryan, L. M., Coull, B. A. (2004): Cholesky residuals for assessing normal errors in a linear model with correlated outcomes. Journal of the American Statistical Association 99466: pg 383-394. Doi 10.1198

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList and PhenTestResult

Examples

file <- system.file("extdata", "test2.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            dataset.colname.weight="Weight.Value",
            testGenotype="Arid4a/Arid4a",
            hemiGenotype="Arid4a/+")
    result <- PhenStat:::testDataset(test,
            depVariable="Cl")
    PhenStat:::qqplotRotatedResiduals(result)

Description

Checks the dependent variable and dataset to make choose the appropriate statistical method. Returns the method or the list of methods that are appropriate for statistical analysis.

Usage

recommendMethod(phenList   = NULL,
                depVariable    = NULL,
                outputMessages = TRUE,
                upper          = 5)

Arguments

Value

Returns the method or the list of methods that are appropriate for statistical analysis.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    PhenStat:::recommendMethod(test,"Lean.Mass")

Description

Method refGenotype returns reference genotype

Value

This function returns reference genotype.

Methods

signature(obj = "PhenList")

signature(obj = "PhenTestResult")

Description

This is an internal function run within RR framework. It performs Reference Ranges Plus test and after that Fisher Exact test on calculated count matrices. As an internal function, it doesn't include extensive error testing of inputs. Please use cautiously if calling directly.

Works with PhenList object created by PhenList function.

Usage

RRTest(
		  phenList,
		  depVariable,
		  outputMessages = TRUE,
		  naturalVariation = 95,
		  controlPointsThreshold = 60
)

Arguments

Value

Returns results stored in instance of the PhenTestResult class

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList and testDataset

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    # "RRTest" function is called from "testDataset" function
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass",
            method="RR")
    # RR test results can be printed out using function "summaryOutput"
    summary(result)

Description

Graph function for the phenotypic dataset. Creates a scatter plot body weight versus dependent variable.

Both a regression line and a loess line (locally weighted line) is fitted for each genotype.

Usage

scatterplotGenotypeWeight(
					  phenList,
					  depVariable = NULL,
					  graphingName = NULL,
					  outputMessages = TRUE
)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    PhenStat:::scatterplotGenotypeWeight(test,
            depVariable="Bone.Mineral.Content",
            graphingName="BMC" )
    file <- system.file("extdata", "test4.csv", package="PhenStat")
    test_1sex <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mysm1/+")
    PhenStat:::scatterplotGenotypeWeight(test_1sex,
            depVariable="Lean.Mass",
            graphingName="Lean Mass")

Description

Graph function for the phenotypic dataset. Creates a scatter plot body weight versus dependent variable. It is based on PhenTestResult object. Both a regression line and a loess line (locally weighted line) is fitted for each genotype.

Usage

scatterplotGenotypeWeightResult(phenTestResult,
                                graphingName = NULL,
                                outputMessages = TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    PhenStat:::scatterplotGenotypeWeightResult(result,
            graphingName="BMC" )
    file <- system.file("extdata", "test4.csv", package="PhenStat")
    test_1sex <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mysm1/+")
    result_1sex <- PhenStat:::testDataset(test_1sex,
            depVariable="Lean.Mass")
    PhenStat:::scatterplotGenotypeWeightResult(result_1sex,
            graphingName="Lean Mass")

Description

Graph function for the phenotypic dataset. Creates a scatterplot split by sex, genotype and batch. refGenotype data points are shown in black and the testGenotype data points are shown in red.

Note: the batches are not ordered with time but allow assessment of how the testGenotype data lie relative to the variation within the refGenotype data.

Usage

scatterplotSexGenotypeBatch(
			  phenList,
			  depVariable = NULL,
			  graphingName = NULL,
			  outputMessages = TRUE
)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    # box plot for dataset with two sexes: males and females
    PhenStat:::scatterplotSexGenotypeBatch(test,
            depVariable="Bone.Mineral.Content",
            graphingName="BMC" )

    file <- system.file("extdata", "test4.csv", package="PhenStat")
    test_1sex <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mysm1/+")
    # box plot for females only dataset
    PhenStat:::scatterplotSexGenotypeBatch(test_1sex,depVariable="Lean.Mass",
            graphingName="Lean Mass")

Description

Graph function for the phenotypic dataset. Creates a scatterplot split by sex, genotype and batch. refGenotype data points are shown in black and the testGenotype data points are shown in red. It is based on PhenTestResult object.

Note: the batches are not ordered with time but allow assessment of how the testGenotype data lie relative to the variation within the refGenotype data.

Usage

scatterplotSexGenotypeBatchResult(phenTestResult,
                                  graphingName = NULL,
                                  outputMessages = TRUE)

Arguments

Value

Empty.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    # box plot for dataset with two sexes: males and females
    PhenStat:::scatterplotSexGenotypeBatchResult(result,
            graphingName="BMC" )

    file <- system.file("extdata", "test4.csv", package="PhenStat")
    test_1sex <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mysm1/+")
    result_1sex <- PhenStat:::testDataset(test_1sex,
            depVariable="Lean.Mass")
    # box plot for females only dataset
    PhenStat:::scatterplotSexGenotypeBatchResult(result_1sex,
            graphingName="Lean Mass")

Description

Method setBatch sets batch column in the dataset

Value

This function sets batch column in the dataset.

Methods

signature(obj = "PhenList")

Description

Method setGenotype sets genotype column in the dataset

Value

This function sets genotype column in the dataset.

Methods

signature(obj = "PhenList")

Description

Method setMissingValue sets missing value in the dataset

Value

This function sets missing value in the dataset.

Methods

signature(obj = "PhenList")

Description

Method setSex sets sex column in the dataset

Value

This function sets sex column in the dataset.

Methods

signature(obj = "PhenList")

Description

Method setWeight sets weight column in the dataset

Value

This function sets weight column in the dataset.

Methods

signature(obj = "PhenList")

Description

This is an internal function run within LR framework. It completes the testing stage of which effects are significant. As an internal function, it doesn't include extensive error testing of inputs. Please use cautiously if calling directly.

It creates start model and modify it after testing of different hypothesis.

The tested effects are:

-batch effect (TRUE if batch variation is significant, FALSE if not), though it is never used in final LR model;

-interaction effect (TRUE if genotype by sex interaction is significant),

-sex effect (TRUE if sex is significant),

-weight effect and variance effect are not tested and used in LR model.

Usage

startLRModel(phenList,
					 depVariable,
					 outputMessages = TRUE,
					 pThreshold = 0.05)

Arguments

Value

Returns results stored in instance of the PhenTestResult class

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "testLR.csv", package="PhenStat")
    testLR <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mock")
    testLR2 <- PhenStat:::LRDataset(testLR,"V2")
    
    # when "testDataset" function's argument "callAll" is set to FALSE 
    # only "startLRModel" function is called - the first step of LR framework
    resultLR <- PhenStat:::testDataset(testLR2,
            depVariable="V2",
            callAll=FALSE,
            method="LR")
    # print out formula that has been created
    PhenStat:::analysisResults(resultLR)$model.formula.genotype
    # print out batch effect's significance 
    PhenStat:::analysisResults(resultLR)$model.effect.batch

Description

This is an internal function run within MM framework. It completes the testing stage of which effects are significant. As an internal function, it doesn't include extensive error testing of inputs. Please use cautiously if calling directly.

It creates start model and modify it after testing of different hypothesis (the model fixed effects).

The model random effects are:

-batch effect (TRUE if batch variation is significant, FALSE if not)

The model fixed effects are:

-variance effect (TRUE if residual variances for genotype groups are homogeneous and FALSE if they are heterogeneous),

-interaction effect (TRUE if genotype by sex interaction is significant),

-sex effect (TRUE if sex is significant),

-weight effect (TRUE if weight is significant).

If user would like to assign other TRUE/FALSE values to the fixed effects of the model then he or she has to define keepList argument which is vector of TRUE/FALSE values.

If user has defined model fixed effects (keepList argument) then function prints out calculated and user defined effects (only when outputMessages argument is set to TRUE), checks user defined effects for consistency (for instance, if there are no "Weight" column in the dataset then weight effect can't be assigned to TRUE, etc.) and modifies start model according to user defined effects.

As the result PhenTestResult object that contains calculated or user defined model fixed effects and MM start model is created.

Usage

startModel(phenList, depVariable,
            equation="withWeight", outputMessages=TRUE,
            pThreshold=0.05, keepList=NULL,modelWeight = NULL,
            threshold = 10^-18,
            check = 1)

Arguments

Value

Returns results stored in instance of the PhenTestResult class

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    # when "testDataset" function's argument "callAll" is set to FALSE
    # only "startModel" function is called - the first step of MM framework
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass",
            callAll=FALSE)
    # print out formula that has been created
    PhenStat:::analysisResults(result)$model.formula.genotype
    # print out batch effect's significance
    PhenStat:::analysisResults(result)$model.effect.batch
    # change the model
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass",
            equation="withWeight",
            callAll=FALSE)
    # print out new formula
    PhenStat:::analysisResults(result)$model.formula.genotype

Description

This is an internal function run within TF framework. It completes the testing stage of which effects are significant. As an internal function, it doesn't include extensive error testing of inputs. Please use cautiously if calling directly.

It creates start model and modify it after testing of different hypothesis.

The tested fixed effects are:

-batch effect (TRUE if batch variation is significant, FALSE if not),

-variance effect (TRUE if residual variances for genotype groups are homogeneous and FALSE if they are heterogeneous),

-interaction effect (TRUE if genotype by sex interaction is significant),

-sex effect (TRUE if sex is significant),

-weight effect (TRUE if weight is significant).

Usage

startTFModel(phenList, depVariable, 
            equation="withWeight", outputMessages=TRUE, 
            pThreshold=0.05, keepList=NULL)

Arguments

Value

Returns results stored in instance of the PhenTestResult class

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test7_TFE.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
                     testGenotype="het",
                     refGenotype = "WT",
                     dataset.colname.sex="sex",
                     dataset.colname.genotype="Genotype",
                     dataset.values.female="f",
                     dataset.values.male= "m",
                     dataset.colname.weight="body.weight",
                     dataset.colname.batch="Date_of_procedure_start")

    test_TF <- PhenStat:::TFDataset(test,depVariable="Cholesterol")
    
    # when "testDataset" function's argument "callAll" is set to FALSE 
    # only "startTFModel" function is called - the first step of TFE framework
    result <- PhenStat:::testDataset(test_TF,
            depVariable="Cholesterol",
            callAll=FALSE,
            method="TF")
    # print out formula that has been created
    PhenStat:::analysisResults(result)$model.formula.genotype
    # print out batch effect's significance 
    PhenStat:::analysisResults(result)$model.effect.batch

Description

Method subsetText returns full name of subset that was analysed by RR or FE framework: Males only, Females only, All.

Value

This function returns full name of subset that was analysed by RR or FE framework.

Methods

signature(obj = "htestPhenStat")

Description

Wrapper for the model fitting results. Returns model fitting and testing results in a user friendly format.

Usage

## S3 method for class 'PhenTestResult'
summary(object, phenotypeThreshold = 0.01,...)

Arguments

Value

Returns model fitting results in a text format for the screen

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

# Mixed Model framework
    file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    summary(result)

    # Fishe Exact Test framework
    file <- system.file("extdata", "test_categorical.csv", package="PhenStat")
    test2 <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Aff3/Aff3")
    result2 <- PhenStat:::testDataset(test2,
            depVariable="Thoracic.Processes",
            method="FE")
    summary(result2)

Description

Wrapper for the model fitting results. Returns model fitting and testing results in a user friendly format.

Usage

summaryOutput(phenTestResult, phenotypeThreshold = 0.01)

Arguments

Value

Returns model fitting results in a text format for the screen

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

# Mixed Model framework
    file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    PhenStat:::summaryOutput(result)
    
    # Fishe Exact Test framework
    file <- system.file("extdata", "test_categorical.csv", package="PhenStat")
    test2 <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Aff3/Aff3")
    result2 <- PhenStat:::testDataset(test2,
            depVariable="Thoracic.Processes",
            method="FE")  
    PhenStat:::summaryOutput(result2)

Description

Statistical analysis manager function in PhenStat package.

Firstly, it performs the checks for dependent variable in the dataset. Some checks are generic, some depends on selected statistical framework/method.

Secondly, if checks are passed it runs the stages of selected framework to analyse dependent variable.

For instance, runs startModel and finalModel for the MM framework if the argument "callAll" is set to TRUE. If framework contains only one stage like in Fisher Exact Test case then runs that one stage regardless "callAll" value.

Usage

testDataset(
          phenList = NULL                ,
          depVariable = NULL             ,
          equation = "withWeight"        ,
          outputMessages = TRUE          ,
          pThreshold = 0.05              ,
          method = "MM"                  ,
          modelWeight = NULL             ,
          callAll = TRUE                 ,
          keepList = NULL                ,
          dataPointsThreshold = 4        ,
          RR_naturalVariation = 95       ,
          RR_controlPointsThreshold = 60 ,
          transformValues = FALSE        ,
          useUnfiltered = FALSE          ,
          threshold = 10 ^ -18           ,
          check = 1                      ,
          upper = 5
)

Arguments

Value

Returns results stored in instance of the PhenTestResult class

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList, plot.PhenTestResult, plot.PhenList

Examples

# Mixed Model framework
    file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    plot(test,depVariable="Lean.Mass")

    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    # print out formula that has been created
    PhenStat:::analysisResults(result)$model.formula.genotype
    summary(result)
    plot(result)






    # Mixed Model framework with user defined effects
    user_defined_effects <- c(keep_batch=TRUE,
            keep_equalvar=TRUE,
            keep_weight=TRUE,
            keep_sex=TRUE,
            keep_interaction=TRUE)
    result3 <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass",
            keepList=user_defined_effects)
    # print out formula that has been created
    PhenStat:::analysisResults(result3)$model.formula.genotype
    summary(result3)
    plot(result3)




    # Fisher Exact Test framework
    file <- system.file("extdata", "test_categorical.csv", package="PhenStat")
    test2 <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Aff3/Aff3")
    plot(test2,depVariable="Thoracic.Processes")
    result2 <- PhenStat:::testDataset(test2,
            depVariable="Thoracic.Processes",
            method="FE")
    summary(result2)
    plot(result2)

Description

Performs diagnostic tests for Logistic Regression model quality of fit. There are no arguments checks assuming that function is called internally from the "finalLRModel" function. Otherwise should be used with precaution.

Usage

testFinalLRModel(phenTestResult)

Arguments

Value

Returns a vector with model fitting results

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

file <- system.file("extdata", "testLR.csv", package="PhenStat")
    testLR <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Mock")
    testLR2 <- PhenStat:::LRDataset(testLR,"V2")
    resultLR <- PhenStat:::testDataset(testLR2,
            depVariable="V2",
            method="LR")
    PhenStat:::testFinalLRModel(resultLR)

Description

Performs diagnostic tests for Mixed Model quality of fit. There are no arguments checks assuming that function is called internally from the "finalModel" function. Otherwise should be used with precaution.

Usage

testFinalModel(phenTestResult)

Arguments

Value

Returns a vector with model fitting results

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    PhenStat:::testFinalModel(result)

Description

Method testGenotype returns test genotype

Value

This function returns test genotype.

Methods

signature(obj = "PhenList")

signature(obj = "PhenTestResult")

Description

Removes of all non-concurrent batches from the dataset. Returns dataset suitable for TF framework or original dataset if data removal was not successful.

Usage

TFDataset(phenList,
          depVariable,
          outputMessages  = TRUE,
          forDecisionTree = FALSE,
          upper           = 5)

Arguments

Value

Returns dataset suitable for TF framework or original dataset if data removal was not successful.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenList

Examples

file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    test2 <- PhenStat:::TFDataset(test,"Lean.Mass")

Description

Method transformation returns transformation values: lamda=value, scaleShift=value

Value

This function returns transformation values.

Methods

signature(obj = "PhenTestResult")

Description

Method transformationJSON returns transformation values as JSON string

Value

This function returns transformation values as JSON string.

Methods

signature(obj = "PhenTestResult")

Description

Method transformationText returns transformation values as one string

Value

This function returns transformation values as one string.

Methods

signature(obj = "PhenTestResult")

Description

Wrapper for the model fitting results. Returns model fitting and testing results in a vector format. Assumes that modelling results are stored in the PhenTestResult object (output from function testDataset).

Usage

vectorOutput(
	phenTestResult,
	phenotypeThreshold = 0.01,
	othercolumns = NULL,
	quote  = '"'
)

Arguments

Value

Returns a vector with model fitting results: model output summary and other values

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

# Mixed Model framework
    file <- system.file("extdata", "test1.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Sparc/Sparc")
    result <- PhenStat:::testDataset(test,
            depVariable="Lean.Mass")
    PhenStat:::vectorOutput(result)

    # Fishe Exact Test framework
    file <- system.file("extdata", "test_categorical.csv", package="PhenStat")
    test2 <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Aff3/Aff3")
    result2 <- PhenStat:::testDataset(test2,
            depVariable="Thoracic.Processes",
            method="FE")
    PhenStat:::vectorOutput(result2)

Description

Returns count matrices from Fisher Exact Test framework in a vector form.

Note: Maximal supported number of dependent variable levels is 10.

Usage

vectorOutputMatrices(phenTestResult,outputMessages=TRUE)

Arguments

Value

Returns a vector with count values.

Author(s)

Natalja Kurbatova, Natasha Karp, Jeremy Mason

References

Karp N, Melvin D, Sanger Mouse Genetics Project, Mott R (2012): Robust and Sensitive Analysis of Mouse Knockout Phenotypes. PLoS ONE 7(12): e52410. doi:10.1371/journal.pone.0052410

West B, Welch K, Galecki A (2007): Linear Mixed Models: A practical guide using statistical software New York: Chapman & Hall/CRC 353 p.

See Also

PhenTestResult

Examples

file <- system.file("extdata", "test_categorical.csv", package="PhenStat")
    test <- PhenStat:::PhenList(dataset=read.csv(file,na.strings = '-'),
            testGenotype="Aff3/Aff3")
    result <- PhenStat:::testDataset(test,
            depVariable="Thoracic.Processes",
            method="FE")  
    PhenStat:::vectorOutputMatrices(result)

Description

Returns TRUE if the weight is in the dataset, FALSE otherwise

Value

TRUE/FALSE

Methods

signature(obj = "PhenList")

signature(obj = "PhenTestResult")