ISU Electrical and Computer Engineering Archives

Suffix Trees and Suffix Arrays in Primary and Secondary Storage

Ko, Pang (2007) Suffix Trees and Suffix Arrays in Primary and Secondary Storage. PhD thesis, Iowa State University.

Full text available as:

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

Abstract

In recent years the volume of string data has increased exponentially, and the speed at which these data is being generated has also increased. Some examples of string data includes biological sequences, internet webpages, and digitalized documents, to name a few. The indexing of biological sequence data is especially challenging due to the lack of natural word and sentence boundaries. Although many algorithms are able to deal with this lack of natural boundaries, they are not able to process the large quantity of data in reasonable time. To speed up the runtime of these algorithms, suffix trees and suffix arrays are routinely used to generate a set of starting positions quickly and/or narrow down the set of possibilities need to be considered. The first contribution of this dissertation is a linear time algorithm to sort all the suffixes of a string over a large alphabet of integers. The sorted order of suffixes of a string is also called suffix array, a data structure introduced by Manber and Myers that has numerous applications in pattern matching, string processing, and computational biology. Though the suffix tree of a string can be constructed in linear time and the sorted order of suffixes derived from it, a direct algorithm for suffix sorting is of great interest due to the space requirements of suffix trees. Our result is one of the first linear time suffix array construction algorithms, which improve upon the previously known O(n log n) time direct algorithms for suffix sorting. It can also be used to derive a different linear time construction algorithm for suffix trees. Apart from being simple and applicable for alphabets not necessarily of fixed size, this method of constructing suffix trees is more space efficient. The second contribution of this dissertation is providing a new suffix tree layout scheme for secondary storage and present construction, substring search, insertion and deletion algorithms using this layout scheme. For a set of strings of total length n, a pattern p and disk blocks of size B, we provide a substring search algorithm that uses O(|p|/B + \log_B n) disk accesses. We present algorithms for insertion and deletion of all suffixes of a string of length m that take O(m \log_B (n+m)) and O(m\log_B n) disk accesses, respectively. Our results demonstrate that suffix trees can be directly used as efficient secondary storage data structures for string and sequence data. The last contribution of this dissertation is providing a self-adjusting variant of our layout scheme for suffix trees in secondary storage that provides optimal number of disk accesses for a sequence of string or substring queries. This has been an open problem since Sleator and Tarjan presented their splaying technique to create self-adjusting binary search trees in 1985. In addition to resolving this open problem, our scheme provides two additional advantages: 1) The partitions are slowly readjusted, requiring fewer disk accesses than splaying methods, and 2) the initial state of the layout is balanced, making it useful even when the sequence of queries is not highly skewed. Our layout scheme, and its self-adjusting variant are also applicable to PATRICIA trees, and potentially to other data structures.

EPrint Type:Thesis (PhD)
Uncontrolled Keywords:String algorithm, suffix tree, suffix array, secondary storage
Subjects:Computer Engineering > SOFTWARE SYSTEMS > Computational Biology and Computational Science
Research Excellence Awards
ID Code:358
Identification Number:Identification Number UNSPECIFIED
Deposited By:Mr. Pang Ko
Deposited On:29 August 2007

Archive Staff Only: edit this record