Interactive R/Shiny application for population genetics analyses
Intertryp · Universite de Montpellier · Cirad · IRD · UCAD
PGA-cmdr covers the full workflow for multilocus, individual-based genotype datasets (microsatellites and other codominant markers): data import and formatting, allele frequencies, general statistics, panmixia and subdivision testing (Weir & Cockerham 1984), genetic diversities, linkage disequilibrium, null allele screening (FreeNA), and isolation by distance — all computed locally, with no data leaving your machine.
PGA-cmdr is distributed through R-universe as ready-to-use binaries for Windows, macOS and Linux: no compiler, Rtools, gfortran or OpenMP setup is needed. All dependencies install automatically.
# 1) Installation (once, or to update to the latest version)
install.packages("pgacmdr",
repos = c("https://kadnaf.r-universe.dev", "https://cloud.r-project.org"))
# 2) Launch the application (every time you use it)
pgacmdr::run_app()
The application opens in your default web browser. To update PGA-cmdr, simply run step 1 again.
Every module follows the same pattern: set parameters, click Run, and
a single .txt file (Allele Freq, Local/Global Panmixia, Subdivision, Diversities, LD) or a .zip of several files (General Stats, Null Alleles, IBD, Mantel) downloads
automatically with every result file needed — nothing is shown on screen
first. Column numbering, loci detection (single- or paired-column allele
format), and output file names are handled automatically wherever possible.
| Module | What it does |
|---|---|
| Import Data | CSV/TXT import (auto-detected comma/semicolon/tab separator, loaded as soon as a file is chosen). Population, Latitude/Longitude column assignment; loci range entered as "number of loci" + "first locus column" (auto-suggested, and automatically adjusted for single- or paired-column allele encoding); a numbered column-reference table is shown so column numbers are never ambiguous. |
| Allele Freq | Per-locus, per-population allele frequencies (plus a global column weighted by sample size), for every population and marker; a single .txt file. |
| General Stats | Per-locus Ho, Hs, Ht and FIS/FST/FIT (Weir & Cockerham 1984), optional Fst-max and Fst' (Meirmans), GST and GST''; per-population Ho, Hs and FIS with Ho/Hs averaged over loci weighted by the number of individuals typed at each locus (Genepop convention); per-locus Ho of every population; per-allele F-statistics. Hs is the unbiased Nei & Chesser (1983) estimator. Output: a .zip of 4 files. |
| Local Panmixia | FIS (Weir & Cockerham 1984) by locus and by population in a single .txt file that states each type of bootstrap and its number of replicates: over individuals (_indiv), over sub-samples (_subs, by locus) and over loci (_loci, Overall row and each population), with two-sided permutation p-values. A second output gives FIS per locus and population with the one-sided heterozygote deficit/excess p-values (FSTAT convention). Fewer than 5 sub-samples or individuals gives NA. |
| Global Panmixia | Multilocus FIT across all populations in a single .txt file: per-locus bootstrap CI over sub-samples, and two Overall rows, Overall (BS/Ss) (sub-samples) and Overall (BS/Loci) (loci); two-sided permutation p-values. |
| Subdivision | A single .txt file with (1) FST per locus (bootstrap over sub-samples, one-sided permutation p-value based on the G statistic) and two Overall rows (over sub-samples, over loci); (2) FIS, FST, FIT, HS, HT multilocus with bootstrap CI over loci and over sub-samples; (3) the FSTAT-style G-based test of genotypic differentiation (p_>=, p_>); optionally the bootstrap replicate values. |
| Diversities | One file, Diversities_confidence_intervals_<date>.txt: HS per locus (bootstrap over individuals and over sub-samples), HS per population (over individuals and over loci), HT per locus, and the multilocus estimators (over loci and over sub-samples); Nei & Chesser (1983) estimators. Intervals from the bootstrap over individuals are shifted by the known downward bias of resampling individuals. Optional bootstrap replicate values. |
| LD | Pairwise linkage disequilibrium between every locus pair (loci, pairs and populations in the order of the data file), per population and combined ("All"), via a permutation-based G-test; a single .txt file. |
| Null Alleles | Null allele frequency estimation per locus × sub-sample (FreeNA EM algorithm, Chapuis & Estoup 2007), raw and ENA-corrected FST and DCSE with bootstrap CI (over loci and over sub-samples), and a full pairwise table (FST, FST-ENA, DCSE, DCSE-INA, linearised FST, geographic distance). |
| IBD | Isolation by distance: Rousset's (1997) regression of linearised genetic distance against geographic distance, and an independent Mantel permutation test (Pearson, Spearman, and Rousset 1D/2D statistics). |
See the package vignette for a full worked example on the bundled Boophilus tick dataset:
vignette("pgacmdr", package = "pgacmdr")
F-statistics follow the unbiased moment estimators of Weir & Cockerham (1984). Gene diversity (Hs, Ht) follows the unbiased estimator of Nei & Chesser (1983). Confidence intervals are obtained by non-parametric bootstrap and p-values by Monte Carlo permutation (10,000 replicates by default in every module, except 5,000 bootstrap replicates in Null Alleles), parallelised in C++ via Rcpp and OpenMP. See the in-app Help tab for full references.
PGA-cmdr: an interactive Shiny application for population genetics data import, exploration, and descriptive analyses. Intertryp / Universite de Montpellier / Cirad / IRD/ UCAD.
Programming: Naffiou Kadiri and Vincent Manzanilla
Conception: Thierry de Meeûs
MIT — see LICENSE.md.