# Similarity Estimation Between DNA Sequences Based on Local Pattern Histograms of Binary Images

## TL;DR

This paper introduces a fast method for comparing DNA sequences using binary images and local pattern histograms, suitable for long sequences.

## Contribution

A novel alignment-free DNA sequence comparison method using local pattern histograms of binary images with linear time complexity.

## Key findings

- The method has linear time complexity, making it practical for long DNA sequences.
- Histogram intersection and Manhattan distance were found to be the most suitable for similarity estimation.

## Abstract

Graphical representation of DNA sequences is one of the most popular techniques of alignment-free sequence comparison. In this article, we propose a new method for extracting features of DNA sequences represented by binary images, in which we estimate the similarity between DNA sequences by the frequency histograms of local bitmap patterns on the images. Our method has linear time complexity for the length of DNA sequences, which is practical even for comparison of long sequences. We tested five distance measures to estimate sequence similarities and found that histogram intersection and Manhattan distance are most appropriate for our method among them.