This revised version of fsthet emphasizes calculating smoothed quantiles instead of bootstrapping. It is faster and can be run on Fst and heterozygosity values calculated by other programs.
The program using these tools is currently under review.
fsthet v1.0
fsthet_1.0.tar.gz
fsthet-vignette.pdf
Please see the vignette for instructions on how to use fsthet.
scripts and programs
scripts
scripts.tar.gz
All of these scripts run the other programs.
numerical analysis
This program runs a numerical analysis as described in Flanagan & Jones "Constraints on the Fst-heterozygosity approach" paper.
It assumes a population size of 1000 individuals per deme, 2000 replications (genes), and 5000 generations.
50 individuals are sampled from each sampled populations.
Sampling either occurs as a certain number of samples per population or by randomly-sampling populations (a single population could be sampled multiple times).
Input
Required parameters
- Base output file name, including the path
- Nm (population size = 1000 times the migration rate)
- The number of demes
- The number of populations to sample
Optional parameters
- Whether random sampling is turned on or off. (Default: random sampling turned on)
- If you want to run it with overdominance, and the overdominance selection coefficient (default: no overdominance)
- If you want to run it with directional selection, and the selection coefficient (default: no directional selection)
Arguments
-o base file name (include path). Example: N1000_s10_
-p Population size (N). This number is set to 1000 if not specified and is constant among all populations.
-n Nm
-d number of demes
-r random sample? In interactive mode, use Y to turn on and N to turn off. In command-line mode, use 1 to turn on.
-s number of populations to sample
-v Overdominance? Follow -v with the selection coefficient (s)
-ds Directional Selection? Follow -ds with the selection coefficient (s)
-h Prints the arguments list
no arguments: interactive mode
Output
- Average Allele frequencies, Fsts, etc. for each generation (*.freqs.txt).
- Heterozygosities, Fsts, and average allele frequency for all demes (not the sampled populations) for each locus at the end of the 5000 generations (*.output.txt).
- Heterozygosities, Fsts, and average allele frequency calculated from the sampled populations (whether populations were randomly sampled or not) for each locus (*.sampledpops.txt).
- Genepop file for the sampled output, ready to be input to LOSITAN.
Installation and Usage
Windows
Put the executable (numerical_analysis.exe) in a useful folder. It is easiest (the path you must provide is shortest) if you put it in your desired output directory.
Ubuntu
Put the executable in a useful folder. It is easiest (the path you must provide is shortest) if you put it in your desired output directory.
**You may need to alter file permissions for it to run:
chmod u+x numerical_analysis
To run the file in interactive mode:
./numerical_analysis
To run the file with arguments:
./numerical_analysis -o file_name_base_ -n 0.1 -d 100 -s 10 -r y
For help:
./numerical_analysis -h
Other operating systems
Compile the source code using the g++ compiler.
For example:
g++ numerical_analysis.cpp -o numerical_analysis -std=c++0x
Note: This program has not been tested on any platforms other than Windows and Ubuntu.
fdist2
This is the fdist2 program from Beaumont & Nichols (1996). It is not mine and belongs to them, but I provide the version I used here so that all of my scripts can be used.