Abstract of Thesis presented at COPPE/UFRJ as a partial fulfillment of the requirements for the degree of Master of Science (M.Sc.)

Multiobjective Evolucionary Algorithms for Biological Multiple Sequences Alignment

Margarita Ramona Ruiz Olazar

April/2007

Advisors:

Eugenius Kaszkurewicz Benjamín Barán Cegla

Department:

Eletrical Engineering

      We studied a methodology for the Biological Multiple Sequences Alignment (MSA) using Multi-Objective Evolutionary Algorithms (MOEAs). This method evolves gradually a given population of alignments, improving the "fitness" of the population measured by two criteria; the quality of the alignment calculated with the "SP score" function using the matrix of substitution BLOSUM62 and the quality of the alignment calculated with "SP score" function using the matrix of substitution PAM250.
      In general, this problem demands a huge processing time; therefore, the proposed implementation tries to benefit from high performance computation using several processors to run a parallel program.
      The main advantage of the proposed methodology is that it can be used for protein and DNA sequences as well. In addition, it gives the possibility of optimizing different objective functions at a time. Experimental results show that MOEAs are efficient methods of optimization and can be used in MSA problems, when the problem domain is determined.