A greedy algorithm的 for optimal recombination.pdfVIP

A Greedy Algorithm for Optimal Recombination Shiquan Wu and Xun Gu Center of Bioinformatics and Biological Statistics Iowa State University, Ames, IA 50011, USA {sqwu,xgu}@iastate.edu Abstract. Let Σ be an alphabet and Σn denote the collection of all sequences of length n over Σ . For any s = a a · · · a a · · · a , s = 1 1 2 j j +1 n 2 b1b2 · · · bj bj +1 · · · bn ∈ Σn , a recombination of s1 and s2 at position j is defined as an operation that crosses s1 and s2 at position j and generates t = a a · · · a b · · · b and t = b b · · · b a · · · a . Denote A and S 1 1 2 j j +1 n 2 1 2 j j +1 n two collections of sequences. In this paper, we discuss generating A from S by a series of recombinations in minimum number of steps. We present a greedy algorithm for finding the optimal recombination evolutionary history from S to any tree A of sequences when |S| = 2. 1 Introduction Various types of mutations on sequences play an important role in computational biology. Transformations using insertion, deletion, substitution, and reversal are widely studied by applying statistics and algorithms (cf. [1,2]). Recently, much attention has been paid to recombination of sequences. Hein developed some algorithms for recombination problems(cf. [3,4]). Kececioglu and Gusfield dis- cussed a recombination distance problem on generating a third sequence from a given pair of sequences in optimal recombination cost (cf. [5]). Generally, an edit distance problem of genomes is widely studied and its aim is to find the optimal evolutionary history from some ancestors to some