Usage: debruijn [-t | -v] <k> <n>
         [<B> <guide function> <neural model>]
         [<neural model #2>] [<neural model #3>] [-eval <SOA>]
Generate random and "path guided" de Bruijn cycles

 Default output has one line of labels, separated by commas
     -t: terse output (no delimiters)
     -v: verbose output in "necklace" format

   k: number of letters (maximum is 36)
   n: length of the word (i.e., level of counterbalance)

 Specify parameters for path-guided cycle
   B: number of bins for available transitions (integer between 1 and k^2)
   guide function: can be specified in three different ways -
                   HRF: guide with power spectrum of the BOLD HRF  -OR-
                   path to .txt file containing guide function -OR-
                   [Tmin,Tmax]: range of the periods of sinusoids
                     (in units of labels) to use as a guide function
   neural model: path to .txt file containing neural model for optimizing
   neural model(#2,#3): optional; detection power only reported

 -eval: Evaluate detection power
   SOA: stimulus-onset asynchrony (in ms)

Detailed help: http://cfn.upenn.edu/aguirre/wiki/public:de_bruijn_software

Reference: GK Aguirre, MG Mattar, L Magis-Weinberg. (2011)
de Bruijn cycles for neural decoding. NeuroImage 56: 1293-1300

Code by: Marcelo Mattar (mattar@sas.upenn.edu)
         Dongbo Hu (dongbo@mail.med.upenn.edu)
         Hamiltonian circuit code from Basil Vandergriend
           (http://webdocs.cs.ualberta.ca/~joe/Theses/vandegriend.html)

v1.6 -- August 09, 2012