A Statistical Similarity/Dissimilarity Analysis of Protein Sequences Based on a Novel Group Representative Vector

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Understanding organismal biology relies on analyzing gene and sequence origins through similarity/dissimilarity.
• Current alignment-free methods generate similarity matrices, which assess sequences individually, potentially limiting comprehensive group analysis.

Purpose of the Study:

• To develop and validate a novel similarity/dissimilarity vector approach for biological sequence analysis.
• To overcome limitations of traditional similarity matrices by representing group characteristics more effectively.

Main Methods:

• Introduced group representatives for three distinct protein sequence sets: beta globin, NADH dehydrogenase subunit 5 (ND5), and spike proteins.
• Calculated a similarity/dissimilarity vector based on these group representatives, replacing the conventional matrix.
• Performed cross-grouping comparisons to confirm the distinctiveness of each protein sequence group.

Main Results:

• The novel similarity/dissimilarity vector approach was successfully applied to beta globin, ND5, and spike protein sequences.
• Cross-grouping comparisons demonstrated the unique characteristics of each analyzed protein group.
• Qualitative comparisons with existing methods and phylogenetic trees validated the utility of the proposed vector approach.

Conclusions:

• The developed similarity/dissimilarity vector method provides a robust alternative to traditional matrices for sequence analysis.
• This approach enhances the understanding of biological relationships and sequence origins within and across groups.
• The method's effectiveness is confirmed by its application to diverse protein families and comparison with established techniques.