Genome classification improvements based on k-mer intervals in sequences

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Vast genomic data necessitates efficient alignment-free comparison methods.
• k-mer based techniques enhance genome sequence comparison accuracy.
• Current methods may lack optimal feature extraction for diverse genomes.

Purpose of the Study:

• To develop a novel k-mer interval-based method for accurate genome sequence comparison.
• To improve feature extraction from genome sequences using k-mer intervals.
• To evaluate the method's performance on various genome datasets.

Main Methods:

• Extracting k-mer intervals as genomic features.
• Calculating sequence distance by comparing k-mer interval distributions.
• Utilizing phylogenetic trees for sequence classification.
• Testing on viral, mitochondrial (MT), microbial, and mammalian genomes.

Main Results:

• The proposed k-mer interval method demonstrates superior classification accuracy compared to existing k-mer based methods.
• The method effectively classifies diverse genome sets.
• Efficient application to long sequences like human and mouse genomes was confirmed.

Conclusions:

• k-mer interval distribution comparison offers a powerful alignment-free genome analysis approach.
• This method provides a significant advancement in genomic sequence comparison and classification.
• The approach is scalable and applicable to large-scale genomic data.