A new distance measure for comparing sequence profiles based on path lengths along an entropy surface

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Comparing DNA sequence profiles is crucial for understanding genomic relationships.
• Existing methods, like comparing consensus sequences, may lack precision for complex structures such as tandem repeats.

Purpose of the Study:

• To introduce a new distance measure for comparing DNA sequence profiles.
• To evaluate the measure's effectiveness in analyzing similarities within tandem repeat families.
• To compare its performance against traditional consensus sequence comparison methods.

Main Methods:

• DNA sequence profiles are represented as character frequency vectors from multiple sequence alignments.
• A novel distance metric is computed based on the minimum path length along an entropy surface.
• Path length estimation involves random graph generation and Dijkstra's algorithm for shortest path calculations.

Main Results:

• The new distance measure was applied to analyze similarities within tandem repeat families in the C. elegans genome.
• The measure demonstrated a more accurate refinement of family relationships compared to consensus sequence comparison.

Conclusions:

• The proposed distance measure offers improved accuracy for analyzing DNA sequence profile similarities, particularly for repetitive elements.
• This method provides a more nuanced understanding of genomic relationships within families of tandem repeats.