Finding Approximate Palindromes in Strings Quickly and Simply

Lloyd Allison,
School of Computer Science and Software Engineering, Monash University, Clayton, Victoria, Australia 3800.
Technical Report 2004/162,
23 Nov. 2004 (draft 19 Sept.)

Abstract:

Described are two algorithms to find long approximate palindromes in a string, for example a DNA sequence. A simple algorithm requires O(n)-space and almost always runs in O(k.n)-time where n is the length of the string and k is the number of ``errors'' allowed in the palindrome. Its worst-case time-complexity is O(n²) but this does not occur with real biological sequences. A more complex algorithm guarantees O(k.n) worst-case time complexity.

 

See [arxiv...pdf][12/'04] & [java code] of the 1st algorithm.