MakeAffectedTrees: Make affected trees
In CoSeg: Cosegregation Analysis and Pedigree Simulation

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

A function which makes trees based on historical data with affection status based on penetrance data.

1 2	MakeAffectedTrees(n = 1, g = 4, gdown = 2, frequencies.df = NULL, demographics.df = NULL, benign.boolean=FALSE)

`n`	An integer giving the number of affected trees to simulate.
`g`	An integer giving the number of ancestral generations of the proband to simulate.
`gdown`	An integer giving the number of descendent generations of the proband to simulate.
`frequencies.df`	A dataframe giving the penetrance of the desired affection. For convenience this can be one of the included penetrance data frames (`BRCA1Frequencies.df`, `BRCA2Frequencies.df`, `MLH1Frequencies.df`), or a user defined penetrance similar to one of those. This defaults to BRCA1Frequencies.df.
`demographics.df`	A dataframe giving the demographics to use. This defaults to `USDemographics.df` which gives historic data on US demographics. Alternatively, the user can choose `ChinaDemographics.df` or input their own dataframe.
`benign.boolean`	A boolean variable which tells the program to simulate disease status for a benign variant when benign.boolean is TRUE.

Returns a tree with affection status. The result is a dataframe with the following columns if using the defaults. If using other penetrance data, column names will differ slightly.

`degree`	An integer giving the generation number of the individual where ancestors have lower values than descendents.
`momid`	A numeric giving the ID of the mother of the individual or NA if the individual is a founder.
`dadid`	A numeric giving the ID of the father of the individual or NA if the individual is a founder.
`id`	A numeric label for the individual.
`age`	A numeric giving the current age of the individual.
`female`	A boolean/integer depicting whether the individual is female(1) or not(0).
`y.born`	An integer depicting the year the individual was born.
`geno`	An boolean/integer giving telling whether the individual is a carrier(1) or not(0).
`famid`	An integer id for the current tree. This is only useful for multiple trees.
`bBRCA1.d`	A boolean/integer depicting whether the individual has breast cancer(1) or not(0).
`oBRCA1.d`	A boolean/integer depicting whether the individual has ovarian cancer(1) or not(0)
`bBRCA1.aoo`	An integer giving the age of onset of the breast cancer if the individual is affected or NA if not.
`oBRCA1.aoo`	An integer giving the age of onset of the ovarian cancer if the individual is affected or NA if not.
`proband`	A boolean/integer labeling the proband(one with the variant and the disease) with a 1 and everyone else with 0. Note that if no suitable proband is found, all members will receive a -1.

Note that CRAN throws up a note when this package is checked saying "no visible binding for global variable...". This is normal and does not affect the analysis.

John Michael O. Ranola and Brian H. Shirts

See also MakeTree, AddAffectedToTree, and PlotPedigree

  ## Not run: 

    #summaries of all the data
    str(BRCA1Frequencies.df)
    str(BRCA2Frequencies.df)
    str(MLH1Frequencies.df)
    str(USDemographics.df)
    str(ChinaDemographics.df)

    #Make a tree with no affection status, g=4 generations above, gdown=2 generations below,
    #seed.age=50, and demographics.df=NULL which defaults to USDemographics.df.
    tree1=MakeTree()

    #Make a tree using Chinese demographics instead.
    tree2=MakeTree(demographics.df=ChinaDemographics.df)

    #Add affection statust to tree2 using BRCA1Frequencies.df which gives the BRCA1
    #penetrance function
    tree1a=AddAffectedToTree(tree.f=tree1,frequencies.df=BRCA1Frequencies.df)

    #make a tree with affection status (same as running MakeTree() and then AddAffectedToTree())
    tree3=MakeAffectedTrees(n=1,g=2,gdown=2,frequencies.df=MLH1Frequencies.df)
    #tree4=MakeAffectedTrees(n=1,g=2,gdown=2,frequencies.df=BRCA2Frequencies.df)


    #Depending on the size of the pedigree generated, probands (defined here as members of the
    #pedigree who are carriers of the genotype with the disease) may not always be present in
    #the pedigree.  To alleviate this problem in this example we manually generate a pedigree.
    #Note that this is from the Mohammadi paper where the Likelihood method originates from.
    ped=data.frame(degree=c(3,2,2,3,3,1,1,2,2,3), momid=c(3,NA,7,3,3,NA,NA,7,NA,8),
      dadid=c(2,NA,6,2,2,NA,NA,6,NA,9), id=1:10, age=c(45,60,50,31,41,68,65,55,62,43),
      female=c(1,0,1,0,1,0,1,1,0,1), y.born=0, dead=0, geno=2, famid=1, bBRCA1.d=0, oBRCA1.d=0,
      bBRCA1.aoo=NA, oBRCA1.aoo=NA, proband=0)
    ped$y.born=2010-ped$age
    ped$geno[c(1,3)]=1
    ped$bBRCA1.d[c(1,3)]=1
    ped$bBRCA1.aoo[1]=45
    ped$bBRCA1.aoo[3]=50
    ped$proband[1]=1

    ped=ped[c(6,7,2,3,8,9,1,4,5,10),]

    #Calculate the likelihood ratio
    CalculateLikelihoodRatio(ped=ped, affected.vector={ped$bBRCA1.d|ped$oBRCA1.d}, gene="BRCA1")

    #Plot the pedigree
    PlotPedigree(ped, affected.vector={ped$bBRCA1.d|ped$oBRCA1.d})

    #Rank and plot the members of the pedigree with unknown genotypes
    RankMembers(ped=ped, affected.vector={ped$bBRCA1.d|ped$oBRCA1.d}, gene="BRCA1")
  
## End(Not run)