MkFDA_HPProtein

 EXAMPLE 6:  Markov Chain FDA for the HP protein model.

0001 % EXAMPLE 6:  Markov Chain FDA for the HP protein model.
0002 
0003  global HPInitConf;   % This is the HP protein instance, defined as a sequence of zeros and ones
0004  HPInitConf =  [zeros(1,12),1,0,1,0,1,1,0,0,1,1,0,0,1,1,0,1,1,0,0,1,1,0,0,1,1,0,1,1,0,0,1,1,0,0,1,1,0,1,0,1,zeros(1,12)]; 
0005  % The number of variables is equal to the sequence length and each
0006  % variables takes values in {0,1,2}
0007  PopSize = 1000; NumbVar = size(HPInitConf,2); cache  = [1,1,1,1,1]; Card = 3*ones(1,NumbVar);   maxgen = 300;
0008  % The Markov chain model(Cliques) is constructed specifying the number of
0009  % conditioned (previous) variables. In the example below this number is
0010  % 1., i.e. p(x) = p(x0)p(x1|x0) ... p(xn|xn-1)
0011  Cliques = CreateMarkovModel(NumbVar,2);
0012  
0013  F = 'EvaluateEnergy'; % HP protein evaluation function
0014  edaparams{1} = {'learning_method','LearnFDA',{Cliques}};
0015  edaparams{2} = {'sampling_method','SampleFDA',{PopSize}};
0016  edaparams{3} = {'repairing_method','HP_repairing',{}}; % Repairing method used to guarantee that
0017                                                         % solutions do not self-intersect
0018  edaparams{4} = {'stop_cond_method','max_gen',{maxgen}};
0019  [AllStat,Cache]=RunEDA(PopSize,NumbVar,F,Card,cache,edaparams) 
0020  
0021  % To draw the resulting solution use function PrintProtein(vector),
0022  % where vector is the best solution found.
0023  vector = AllStat{maxgen,2}
0024  PrintProtein(vector)