Similarity Detection in Biological Sequences using Parameterized Matching and Q-gram

Abstract

Whenever characterization of a new DNA sequence takes place then, database search is carried out to find whether homolog’s of gene is present or not. Various evolutions in this field have marked the shift from exact matching to a completely different concept, parameterized matching. Parameterized matching is detected by consistent renaming of text and pattern using bijective mapping. While finding matches between pattern and text, the PBMH-Hash algorithm results in frequent occurrence of false matches with large number of character comparison. This paper presents a new algorithm to detect similarity in biological sequences. The proposed algorithm is based on the concept of Berry-Ravindran algorithm and q-Gram. Analysis shows that our algorithm outperforms existing PBMH-Hash algorithm.