Question: Fdist2 via Arlequin
1
gravatar for remi.b.md
21 months ago by
remi.b.md60
remi.b.md60 wrote:

Goal

I am trying to use the FDist2 methodology. Lositan seems to not be working anymore. Arlequin has an implementation of FDist2, so I am now trying to use FDist2 via Arlequin via the command line.

Note that I am on MAC OSX. If I use the command line, I can also do it on Linux.

Issue

I will have to specify in the .ars file that I want Fdist2 to be performed. Each specific requirement has a keyword that has to be associated to a value (typically o or 1) in the .ars file. For example ComputeStandardDiversityIndices=1 and LocByLocAMOVA=1 can be set to 1 as shown if we want these specific statistics to be computed. In the arlsumstat_readme.txt, there is a list of keywords called Keywords in ssdefs.txt (I don't really know what ssdefs.txt is) but these keywords don't match those used in the example files found in the .zip file for Arlecore (typically even AMOVA is absent) and none of these keywords seem to be described as what I want to do. I can't find any of these keywords in the manual either.

In the .zip file for Arlecore, there are a number of examples, incl. one called detSel which sounds exactly like what I would like to do but for some reason the .ars file is absent!

Questions

My questions are

  • Where can I find these keywords?
  • More specifically, what is the keyword for using Fdist2?
  • Could you provide me with a reproducible example of a FDist2 analysis?
ADD COMMENTlink modified 21 months ago • written 21 months ago by remi.b.md60
1
gravatar for remi.b.md
21 months ago by
remi.b.md60
remi.b.md60 wrote:

Facing similar issues, Whitlock and Lotterhos 2014 have written their own implementation of Fdist2. It is available via this Dryad repository.

It is not quite really user friendly but it comes with a R wrapper code that is nicely commented.

ADD COMMENTlink written 21 months ago by remi.b.md60
1
gravatar for h.mon
21 months ago by
h.mon24k
Brazil
h.mon24k wrote:

Not what you asked, but I could run Lositan using command-line javaws from a Java 1.7 custom install on Linux.

  1. install Java 1.7 SDK (JRE may work, I didn't test) to custom location on $HOME
  2. set PATH and JAVA_HOME
  3. run ControlPanel (needed GUI here), go to security tab, Exception Site List, add Lositan site ( http://popgen.net/soft/lositan/code2/ )
  4. run javaws http://popgen.net/soft/lositan/code2/selwb4.jnlp
  5. Click Later when asked to update Java
  6. Click I accept the risk and want to run this application box, then run.
ADD COMMENTlink written 21 months ago by h.mon24k

Again not what you asked, but did you consider BayeScan?

ADD REPLYlink written 21 months ago by h.mon24k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1140 users visited in the last hour