This paper describes a general method for controlling the running time of similarity search algorithms. Our method can be used in conjunction with the seed-and-extend paradigm employed by many search algorithms, including BLAST. We introduce the concept of a seed tree, and provide a seed tree-pruning algorithm that affects the specificity in a predictable manner. The algorithm uses a single parameter to control the speed of the similarity search. The parameter enables us to reach arbitrary levels between the exponential increases in running time that are typical of seed-and-extend methods. © Springer-Verlag 2004.
CITATION STYLE
Csúrös, M. (2004). Performing local similarity searches with variable length seeds. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3109, 373–387. https://doi.org/10.1007/978-3-540-27801-6_28
Mendeley helps you to discover research relevant for your work.