RankMembers: A function to calculate the likelihood ratio
In CoSeg: Cosegregation Analysis and Pedigree Simulation

Description Usage Arguments Value Author(s) Examples

This function calculates the likelihood ratio for an allele causing a disease asssuming that the allele is extremely rare so that all family members who have the allele got it directly from a common ancestor in the pedigree.

1 2	RankMembers(ped, affected.vector, gene="BRCA1", legend.location="topleft", legend.radius=0.1)

`ped`	A dataframe of a pedigree containing id, momid, dadid, age(Current age or age affected), y.born(year born), female(a logical where 1 is female and 0 is male), and genotype(0 is non-carrier, 1 carrier, and 2 unknown) in any order but with these specific column names.
`affected.vector`	A vector giving the affection status of the individual. 0 is unaffected, 1 is affected, and 2 is unknown.
`gene`	A text string (e.g. BRCA1, BRCA2, or MLH1) giving the gene name for analysis.
`legend.location`	A string indicating the placement of the legend. These can be "topleft", "topright", "bottomleft", "bottomright".
`legend.radius`	A real number indicating the size of the legend.

`likelihood.ratio`	The resultant likelihood ratio
`separating.meioses`	The number of meioses separating all individuals known to have the genotype including those that are are indirectly known to have the genotype.
`number.genotypes.found`	The number of permissible genotypes found for the pedigree.

John Michael O. Ranola and Brian H. Shirts

  ## Not run: 
    #Load all the data included in the CoSeg package.
    data(BRCA1Frequencies.df, package="CoSeg")
    data(BRCA2Frequencies.df, package="CoSeg")
    data(MLH1Frequencies.df, package="CoSeg")
    data(USDemographics.df, package="CoSeg")
    data(ChinaDemographics.df, package="CoSeg")

    #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)