#version 0.5-10 16.11.2021
-added functions pi.dosage, thetaWatt.dosage and TajimaD.dosage to calculate these quantities from dosage data
-fixed a bug in write.ped when the argument pop!=NULL, the first column contains the genotypes at the first locus
-fixed a bug in genet.dist, it now output the names of the populations rather than a number for some of the options
-beta.dosage can now be estimated from a matching matrix, with argument MATCHING=TRUE


#version 0.5-9 22.02.2021
-added function Hs and Ho to produce mean population gene diversities and observed heterozygosity respectively
-created function matching, which estimates matching or allele sharing between pairs of inividuals
-modified beta.dosage and fs.dosage to use matching
-fs.dosage now also estimates pairwise fst
-modify plot.fs.dosage: replaced barplots of Fis with image of pairwise Fsts
-created alias pairwise.fst.dosage
-fixed a bug in getal when pop is a factor, on pop id was 1 but not first level in factor
-fixed a bug in getal.b, which was removing the first column of the dataframe (argument to getal.b is a data frame containing genotypes only)

#version 0.5-8 09.08.2020 
-created class betas for objects generated by function betas
-new function pairwise.betas for pairwise betas FST
-new function boot.ppbetas for bootstrap ci of pairwise betas FSTs


#version 0.5-04 17.07.2020
-new function read.VCF reads VCF files

#version 0.5-02 16.07.2020
-new function fs.dosage, which estimates individual inbreeding coefficients, population Fis, population Fst (beta^X) and beta^{XY} with a plot method. Can use matching proportion rather than dosages (faster)
-handling of missing data for dosage format 
-conversion to bed format from ms using ms2bed
-updated vignettes


#version 0.04-30 21.09.2019
-new function fst.dosage
-new functions kinship2grm,kinship2dist,grm2kinship,kinshipShift

#version 0.04-29 25.05.2018
-new function beta.dosage, to estimate kinship from allelic dosage matrix
-improved mat2vec and vec2mat

#version 0.04-27 21.01.2017
-new features in betas, now can produce individual coancestries
-new function to import genotypes from Hudson' ms

#version 0.04-25 31.03.2016
-fixed bug in genind2hierfstat when the number of digits for allele encoding varies

#version 0.04-24 22.02.2016
-fixed bug in pop.freq when the number of alleles per locus was 9
-Simplified genind2hierfstat (Eric Archer contrib)

#version 0.04-23 07.12.2015
-fixed bug in basic.stats when dealing with haploid data
-included URL to references in description file

#version 0.04-22 04.12.2015
-put DOI in between <>

#version 0.04-21 04.12.2015
-added DOI of yang paper in DESCRIPTION file

#version 0.04-20 24.11.2015
-fixed bug in pairwise.neifst when population is a factor
-New function genind2hierfstat to read genind objects
-several hierfstat functions (basic.stats, wc,...) now read genind objects from adegenet
-new function to write ped files (plink) and bayescan files
-new function for pairwise fst:  pairwise.WCfst 


#version 0.04-18 16.10.2015
-fixed bugs in plot.indpca, genet.dist
-added a function to compte corrected Assignment indexs (AIc)
-added a function to test for sex biased dispersal
-added a function to read output from quantinemo
-added an example output quantinemo file

#version 0.04-16 14.09.2015
-fixed several bugs
-finite and non equilibrium island model can be generated with sim.genot.t
-added pairwise FST (Nei87 and WC84) as geentic distances in genet.dist


#version 0.04-15 30.09.2014
-fixed a bug in basic.stats, when only one population with index 2, caused a non conformable array error message. 
Now add 1 to the last element of the population vector

#version 0.04-14 25.09.2014
-Modified sim.genot to allow for different island sizes and independant inbreeding coefficient

#version 0.04-12 06.02.2014
-caught bugs in getal and getal.b relative to number of digits encoding alleles (1001-> modulo 1000 not 100)


#version 0.04-11 31.10.2013
-fixed a bug in boot.ppfst which stop the looping over populations to be carried out correctly
-basic.stats can now take input from data set made of one sample only (a dummy extra line is added to the data set)

#version 0.04-10 09.04.2013
-changed the help file for read.fstat and read.fstat.data so that the examples now call find.package rather than .find.package
-added function to write fstat files, structure files, and sub-sampled individuals from populations
-changed the printout of basic.stats, now an object
-created a function (requiring ade4) indpca carrying out a PCA on individual allele frequencies
-Fixed various other small bugs


#version 0.04-9 04.10.2012
modified bs and wc to allow for haploid data and for better output
#version 0.04-8 30.07.2012
modified getal and pp.fst to allow for non continuous numbering of populations
#version 0.04-7 26.07.2012
added beta.r for estimation of WH02 beta
#version 0.04-6 23.11.2011
cleaned up things to avoid warning messages
#version 0.04-5 30.07.2010
changed \format to \value in Rd file
#version 0.04-4 15.08.2006
removed data frame names (names.data<-names(data)) in prepdata, caused a bug from R version 2.3.1 when only one level and does not seem to do anything


#version 0.04-3 30.06.2006
fixed a bug in boot.vc, now the function removes monomorphic loci before carrying out the bootstrap, and gives an error message if the number of polymorphic loci is less than 5

#version 0.04-2

fixed a bug in genot2al when alleles were coded with 3 digits but with number not exceeding 9

#version 0.04-1

boot.vc now provides bootstrap CI for variance components.

#version 0.03-4

fixed bug in samp.within, now it can handle lev with only one subgroup

#version 0.03-3

Again fixed bugs in test.between, test.between.within and test.within. replaced as.integer(x) with as.integer(factor(x)) forcing consecutive numbering of factor levels. 

#version 0.03-2
Again fixed bugs in test.between, test.between.within and test.within. No renumbering of factor levels, factors are just forced as integers and sorted once.

updated some references

#version 0.03-1
in varcomp.glob, added diploid as a parameter passed to varcomp. Absence of this option caused the function to report an error message when the data were haploid
Changes were made to functions test.between, test.between.within and test.within. The vector rand.unit and test.lev are transformed such that the different units are numbered consecutively
Updated the reference list

#version 0.02.2
in samp.between and samp.within, changed x<-NULL (which caused message Error: more elements supplied than there are to replace) into x<-list().  

#version 0.02

The function test.between, test.between.within and test.within now sort the data before doing the resampling. This avoids errors when the vector within, rand.unit and test.lev are not sorted. 

In function test.between and test.between.within, the vector rand.unit is transformed such that the different units are numbered consecutively

