Applying RECPARS

For our analysis, we used the program written by Kim Fisker , which improves on Jotun Hein's original version. Fisker's program reads in a file with various options, which we have set as follows:

substfile = parameters.recpars
datafile = dna.recpars
rekombcost = NNNN
initial_phyl = best 0 100
rounds = 10

• substfile is a file with substitution costs, which were chosen uniformly:

  	A	C	G	T	-
  A	0	10	10	10	10
  C	10	0	10	10	10
  G	10	10	0	10	10
  T	10	10	10	0	10
  -	10	10	10	10	0

• datafile
  is the file with the DNA sequence alignment, generated with SIRENS.
• recombcost
  is the cost of a recombination event in the parsimony analysis. We varied this value between 10 and 100, corresponding to a variation of the recombination cost/substitution cost ratio between 1 and 10.
• initial_phyl
  specifies which segment in the sequences should be used for finding the initial phylogeny. We used Kim Fisker's default value of the first 100 positions.
• rounds
  is the number of parses over the sequences. The default value is 2 and means that the sequences are scanned first forward and then backward. We increased this value to 10 to increase the accuracy of the method, although this did not make any difference.

Back to the main page