startTFModel: Method "startTFModel"
In PhenStat: Statistical analysis of phenotypic data

Description Usage Arguments Value Author(s) References See Also Examples

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).

1
2
3

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

`phenList`	instance of the `PhenList` class; mandatory argument
`depVariable`	a character string defining the dependent variable of interest; mandatory argument
`equation`	a character string defining the equation to use. Possible values "withWeight" (default),"withoutWeight"
`outputMessages`	flag: "FALSE" value to suppress output messages; "TRUE" value to show output messages ; default value TRUE
`pThreshold`	a numerical value for the p-value threshold used to determine which fixed effects to keep in the model, default value 0.05
`keepList`	a logical vector defining the significance of different model effects: keep_batch, keep_equalvar, keep_weight, keep_sex, keep_interaction; default value NULL

Returns results stored in instance of the PhenTestResult class

Natalja Kurbatova, Natasha Karp, Jeremy Mason

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.

PhenList

    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