compare.clines: Compare Clines
In introgress: methods for analyzing introgression between divergent lineages

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

This function contrasts patterns of introgression between two hybrid zones.

1 2	compare.clines(cline.data1=NULL,cline.data2=NULL,sig.test=FALSE, n.reps=1000)

`cline.data1`	a list that is the product of the `genomic.clines` function.
`cline.data2`	a list that is the product of the `genomic.clines` function.
`sig.test`	a logical specifying whether to perform significance testing.
`n.reps`	numeric value specifying the number of permutations to conduct for signficance testing.

This function estimates that likelihood of the count data from cline.data2 given the regression models from cline.data1 and cline.data2 and returns the log ratio of the latter to the former. cline.data1 and cline.data2 are lists returned by genomic.clines each based on a different hybrid zone between the same parental populations or species; the molecular markers and parental allele frequencies should be the same for both hybrid zones. The range of hybrid index estimates for individuals comprising cline.data2 should exceed the range of hybrid index estimated for individuals comprising cline.data1 to avoid predicting allele counts using the regression model from cline.data1 beyond the range of hybrid indexes that were included in the original model. The log likelihood ratios returned by this function can be used to determine the degree of congruence for marker specific patterns of introgression between the analyzed hybrid zones. This function includes a permutation procedure for testing whether log likelihood ratios are significantly greater than expected by chance, which is invoked by setting sig.test = TRUE. This option should only be employed when both hybrid zones contain many individuals with low and high hybrid index estimates to minimized the problem described above.

See Gompert and Buerkle (2009a, 2009b) for additional details and examples.

A matrix with log likelihood ratios for each marker and P-values from significance testing if sisg.test = TRUE.

Zachariah Gompert zgompert@uwyo.edu, C. Alex Buerkle buerkle@uwyo.edu

Gompert Z. and Buerkle C. A. (2009) A powerful regression-based method for admixture mapping of isolation across the genome of hybrids. Molecular Ecology, 18, 1207-1224.

Gompert Z. and Buerkle C. A. (2009) introgress: a software package for mapping components of isolation in hybrids. Molecular Ecology Resources, in preparation.

genomic.clines, prepare.data, est.h

## Not run: 
## load simulated data
## markers have fixed differences, with
## alleles coded as 'P1' and 'P2'
data(AdmixDataSim1)
data(LociDataSim1)

## use prepare.data to produce introgress.data
introgress.data<-prepare.data(admix.gen=AdmixDataSim1,
                               loci.data=LociDataSim1,
                               parental1="P1", parental2="P2",
                               pop.id=FALSE, ind.id=FALSE, fixed=TRUE)

## estimate hybrid index
hi.index<-est.h(introgress.data=introgress.data,
                loci.data=LociDataSim1, p1.allele="P1",
                p2.allele="P2")

## random sampling to divide data into two sets of 100 individuals,
## this creates two admixed populations (hybrid zones)
numbs<-sample(1:200,200,replace=FALSE)
sam1<-numbs[1:100]
sam2<-numbs[101:200]

## estimate genomic clines for each data set,
## significance testing is not conducted
clines.out1<-genomic.clines(introgress.data=introgress.data,
                            hi.index=hi.index,loci.data=LociDataSim1,
                            sig.test=FALSE, ind.touse=sam1)

clines.out2<-genomic.clines(introgress.data=introgress.data,
                            hi.index=hi.index,loci.data=LociDataSim1,
                            sig.test=FALSE, ind.touse=sam2)

## compare clines between data sets, with significance testing
comp.out<-compare.clines(clines.out1,clines.out2,sig.test=TRUE,
                         n.reps=1000)

write.table(comp.out, file="compareClines.txt",
            quote=FALSE, sep=",")

## End(Not run)