This tutorial covers the use of the Barrier 2.2¹ software. Barrier is a GUI software that implements the Monmonier’s maximum difference algorithm² to determine if there are any geographic barriers to gene flow on a given landscape. Barrier uses a genetic distance matrix (e.g., Nei's) and geographic data as input (described below).

This tutorial will be using raw microsatellite data³ downloaded from Dryad⁴ to predict where potential geographic barriers are between populations of pawpaws (Asimina triloba). This dataset includes two population types: wild [62 populations] and anthropogenic [20 populations]. This tutorial will only be using the anthropogenic populations since no locality data was provided for the wild populations. The dataset containing only the anthropogenic populations is called 'pawpaw_AnthroData.xlsx'.

For this tutorial, we will be using a Nei’s unbiased genetic distance matrix as the genetic distance matrix input. That matrix was generated using GenAlEx 6.5 ^5,6 in Microsoft Excel (2012). Click on the "Frequency-Based" tab then click "Frequency...". A pop up window should appear asking for "Allele Frequency Data Parameters":

Input your data for the #Loci, #Samples, #Pops, and individual Pop. Sizes. We will be working with codominant data, so that will be marked as well under "Data Format", although that can change depending on your own study. Next, a "Codominant Frequency Options" window will pop up. Since we only need Nei's unbiased genetic distance matrix, only that box is clicked. Afterwards, we click ok.

From there, a new tab should appear called "uNeiP" which contains the genetic distance matrix we need.

The geographic data was obtained from Table 1 in Wyatt et al. (2021)³.

Both resulting matrixes will need to be changed to .txt files as input to Barrier. An example of the input data files used in this tutorial, named ‘Asimina_triloba_geo_dist.txt’ and ‘Asimina_triloba_Neis_dist.txt’, can be found here.

The Barrier software used to be available for download here; however, it doesn't appear to work from that site anymore anymore. Therefore, I have uploaded the software and manual for you to use here.

NOTE: Some people have had issues with file format on their computers when not on others. If that is the case, please send your files to Erika Moore ([email protected]) and she can test it using her version and operating system (Windows 10). It may be a relict of not downloading from the source.

Once Barrier is downloaded, you can open the GUI software. From my own experience, I have always had to "Run as Administrator". If you have issues opening the software, give that a try.

Next, you need to start a new analysis. To do that, you will you go to File -> New... -> /User/BarrierFolder/Asimina_triloba_geo_dist.txt. The software may not show a preview of your file/data. If that happens, click around in the "Data Setup" section to find the data. Once it's found, it will look something like this:

The result will show different points that are connected by lines:

Next, click on Barriers -> New...-> Load matrix data... -> /User/BarrierFolder/Asimina_triloba_Neis_dist.txt. You may have to find your data again. Once you do though, the screen will look something like this:

Click ok.

Then you can designate the number of barriers and get a summary report. I did 10 possible barriers. This was my result:

The summary report ("Asimina_triloba_SummaryReport10.TXT") can be found here. The other barriers can be tested and different levels to see the order of which Barriers were found.

The results of our Barrier analysis shows that the first Barrier present in the analysis is between population 6(LCN) with populations 3(CON), 10(MC1), 12(MC3), 7(LM1), 8(LM2), and 5(FHM). If placed on a map, we may be able to find natural or artificial barriers to gene flow such as mountains, rivers, cities, etc.

[1]: Manni, F., Guérard, E., and Heyer, E. (2004). Geographic patterns of (genetic, morphologic, linguistic) variation: How barriers can be detected by using Monmonier’s algorithm. Human Biol, 76: 173–190.

[2]: Monmonier, M.S. (1973). Maximum-Difference barriers: An alternative numerical regionalization method. Geogr. Analysis, 5: 245–261.

[3]: Wyatt, G.E., Hamrick, J.L., and Trapnell, D.W. (2021). The role of anthropogenic dispersal in shaping the distribution and genetic composition of a widespread North American tree species. Ecology and Evolution, 11(16): 11515–11532. https://doi.org/10.1002/ece3.7944.

[4]: https://doi.org/10.5061/dryad.5x69p8d3g.

[5]: Peakall, R. & Smouse, P.E. (2006). GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes, 6: 288-295. https://doi.org/10.1111/j.1471-8286.2005.01155.x.

[6]: Peakall, R. & Smouse, P.E. (2012). GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research--an update. Bioinformatics (Oxford, England), 28(19), 2537–2539. https://doi.org/10.1093/bioinformatics/bts460.

[7]: RStudio Team. (2020). RStudio: Integrated Development for R. RStudio, PBC, Boston, MA. http://www.rstudio.com/.