GenPopPoly tab
This tab allows users to compute a list of population genetic indices
suitable to analyse genetic diversity and population structure of
polyploid populations with a special focus on reproductive modes. These
indices are useful and efficient to estimate rates of clonality,
autogamy (selfing) and allogamy on genotypes of sampled populations
sampled at one time (Castric et al. 2002, David et al. 2007, Hardy 2016,
Stoeckel et al. 2021). Users select the population(s) to be analysed,
select the analyses to be computed and reported, launch the computation
and can directly browse the results for a first sight in the integrated
calc viewer. The results are also saved in a text-file (separator
tabulation) in the folder containing GenAPoPop executable.
Result files can readily be opened by all spreadsheet applications to be
explored and manipulated to do tables and figures. The output file
presents first all intra-population indices computed per population,
then computed overall populations. It includes genotypic and genetic
diversity indices as recommended in Stoeckel et al. (2021),
probabilities of identity for diploids and autopolyploids (Jacquard
1970, Evett & Weir 1998, Waits et al. 2001, Huang et al. 2015), the
four first moments (i.e., mean, variance, skewness and kurtosis)
of inbreeding coefficient FIS in populations (Stoeckel
& Masson 2014). It also provides a list of multi-locus genotypes
(commonly named MLG in literature or genet) with their shared genotype,
and in the last column, the number of repeated genotypes (ramet) found
in the considered population. In each and overall populations, it
reports genotypic diversity indices including the index of clonal
diversity (R , Dorken & Eckert, 2001) and the size distribution
of lineages (D* of Simpson and Pareto 𝛽 , Arnaud-Haond et al.,
2007) computed properly for autopolyploids. We deliberately discarded
many other indices to help users robustly interpreting genotypic
diversity in their populations. Despite Pareto 𝛽 is far more
robust than the R to assess genotypic diversity in sampled
populations (Stoeckel et al. 2021, Arnaud-Haond et al. 2020), we still
compute R for reference, as this one was historically massively
reported in past literature. The output also provides the mean
correlation coefficient of genetic distances between unordered alleles
at all loci, usually named \({\overset{\overline{}}{r}}_{d}\) as an
overall measure of linkage disequilibrium per population and overall
populations (Agapow & Burt, 2001). This index, ranging from slightly
negative or 0 (no correlation) to 1 (maximum association of alleles over
all loci), presents the advantage of limiting the dependency of the
correlation coefficient on the number of alleles and loci.
GenAPoPop also provides per population and overall populations
a table of classical intra-populational genetic indices per locus:
observed heterozygosity, raw and unbiased expected heterozygosity (also
name gene diversity), resulting raw and unbiased inbreeding coefficient
(Fis ) accounting for intra-individual genetic variation as a
departure from Hardy-Weinberg assumptions of the genotyped populations
and the raw and effective number of alleles (Ae , Weir 1996).
On a side and more experimental
part, GenAPoPop allows computing analysis of molecular variance
(AMOVA) computed following Meirmans & Liu (2018) and Weir (1996)
equations and recommendations, including the Fis, Fst and Fit per
population, over all populations, per marker and over all markers. These
results can already be obtained using Polygene and Genodive.
GenAPoPop also provides in this section the overall and
pairwise-population rhost . rhost measures the genetic
differentiation between populations as the Fst value that would
have the same haploid population sizes connected with the same migration
rate, and present the advantage to be comparable between species and
populations of different ploidy levels (Ronfort et al. 1998, Meirmans &
Van Tienderen 2013). These indices of genetic
differentiation/structuration are a good complement to the minimum
spanning tree of the genetic distances between individuals when coloured
or tagged by population to get a picture of the genetic structure of
genotyped populations (see below). As these indices are also computed in
Spagedi and Polygene, we invite users to also compare
their results with these softwares.
GenAPoPop was thinked and designed to complement
Genodive, Polygene that performs hierarchical,
Bayesian clustering and parentage analysis, and Spagedi that
already performs multiple spatial analyses and that can be used to
estimate selfing rate. In this tab, GenAPoPop users can export
automatically their datasets in a Spagedi-format file that will
be recorded in the same folder under the same imported data name extend
with “_spagedi_ready.txt ”. This file that can be easily
imported to extend and access complementary analyses in the previously
cited software, including Spagedi and Polygene, and we
greatly encourage future GenAPoPop users to analyse their data
with multiple softwares to get the most complete view of their dataset.