Fast and accurate genome comparison using genome images: The Extended Natural Vector Method

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Genome comparison is crucial in bioinformatics.
• Chaos Game Representation (CGR) is an effective method for mapping genome sequences into images.
• Extended Natural Vectors (ENVs) can be derived from CGR images based on intensity distributions.

Purpose of the Study:

• To develop a novel, efficient, and accurate method for DNA sequence comparison and phylogenetic analysis.
• To establish a one-to-one correspondence between genome sequences and their Extended Natural Vectors (ENVs).
• To evaluate the performance of the CGR-ENV method against existing techniques like Multiple Sequence Alignment (MSA).

Main Methods:

• Genome sequences are converted into Chaos Game Representation (CGR) images.
• Extended Natural Vectors (ENVs) are generated from CGR images, capturing intensity distribution.
• The distance between DNA sequences is defined by the distance between their corresponding ENVs.
• Phylogenetic trees are constructed by clustering and classifying datasets using the CGR-ENV method.

Main Results:

• The CGR-ENV method demonstrates favorable classification accuracy and efficiency compared to MSA and other alignment-free methods.
• Successful construction of phylogenetic trees for various datasets, including Influenza A viruses, Bacillus genomes, and Conoidea mitochondrial genomes.
• The ENV mapping provides a unique vector for each CGR image, enabling precise sequence comparison.

Conclusions:

• The CGR-ENV method is a powerful tool for efficient DNA comparison and phylogenetic studies.
• This approach offers significant insights into genome evolution and comparative genomics for large datasets.
• The study highlights the potential of numerical methods in advancing bioinformatics and evolutionary biology research.