FSTAT

version 2.9.3.2 (Feb. 2002)
FSTAT estimates and tests gene diversities and differentiation statistics from codominant genetic markers. It computes both Nei and Weir & Cockerham families of estimators of gene diversities and F-statistics, and tests them using randomisation methods. Jackknife and Bootstrap confidence intervals are also provided. The windows version also estimates statistics developed for the stepwise mutation model (Rst and the like)

• Windows (95 and above)
• Macintosh with PC emulation software

• Microsatellites
• Standard (multi-allelic markers)

input file of format GENEPOP can be translated to a FSTAT input file in the file conversion menu

• named FILENAME.DAT
• total number of individuals in the data set needs to be less than 20’000
• first line: contains 4 numbers separated by any number of spaces:
  • number of samples, np (‹=200)
  • number of loci, nl (‹=100)
  • highest number used to label an allele, nu (‹=999)
  • a 1 if the code for alleles is a one digit number (1-9), a 2 if code for alleles is a 2 digit number (01-99) or a 3 if code for alleles is a 3 digit number (001-999)
• followed by nl lines:
  • each containing the name of a locus, in the order they will appear in the rest of the file
  • On line nl+2, a series of numbers as follow: 1 0102 0103 0101 0203 0 0303
    • first number: identifies the sample to which the individual belongs
    • second: the genotype of the individual at the first locus, coded with a 2 digits number for each allele
    • third: the genotype at the second locus, until locus nl is entered (in the example above, nl=6)
  • Missing genotypes are encoded with 0 (0001 or 0100 are not a valid format, so both alleles at a locus have to be known, otherwise, the genotype is considered as missing)
  • no empty lines are needed between samples
  • number of spaces between genotypes can be anything
  • Numbering of samples need not be sequential
  • the number of samples np needs to be the same as the largest sample identifier
  • Samples need not to be ordered
  • nu needs to be equal to the largest code given to an allele (even if there are less than nu alleles)

• code for alleles is a one digit number:

  3  5  4  1
  loc-1
  loc-2
  loc-3
  loc-4
  loc-5
     1   44 43 43 33 44
     1   44 44 43 33 44
     1   44 44 43 43 44
     2   44 44 33 32 44
     2   44 33 44 43 44
     2   44 43 44 43 44
     3   44 44 44 43 44
     3   44 44 44 44 44
     3   44 44 43 21 44

• code for alleles is a two digit number:

  3  5  4  2
  loc-1
  loc-2
  loc-3
  loc-4
  loc-5
  1       0404 0403 0403 0303 0404
  1       0404 0404 0403 0303 0404
  1       0404 0404 0403 0403 0404
  2       0404 0404 0303 0302 0404
  2       0404 0303 0404 0403 0404
  2       0404 0403 0404 0403 0404
  3       0404 0404 0404 0403 0404
  3       0404 0404 0404 0404 0404
  3       0404 0404 0403 0201 0404

text file containing names (labels) for the population samples:

• extension: .LAB
• Each line should contain the name (label) of one sample
• samples should appear in the same order as in the .DAT file
• labels can be any length but will be truncated to six characters in the output files
• example:

  Stade de France
  Twickenham
  Arms Park

No specific extension is required for this file

• example:

         8	     3
  var0 (YANOM.GEN)
  var1 (YANOM.ANT)
  var2 (YANOM.GEO)
  35.00000	96.00000	 9.00000	
  44.00000	147.00000	28.00000	
  38.00000	88.00000	20.00000	
  47.00000	295.00000	152.00000	
  56.00000	318.00000	161.00000	
  59.00000	309.00000	178.00000	
  52.00000	284.00000	149.00000	
  65.00000	339.00000	158.00000	

Goudet, J. 2001. FSTAT, a program to estimate and test gene diversities and fixation indices (version 2.9.3). Available from http://www2.unil.ch/popgen/softwares/fstat.htm. Updated from Goudet (1995)