Min-frame transformation enables more sensitive viral genome alignment.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Maximal unique matches (MUMs) are crucial for genome comparison and multiple genome alignment.
• Exact string matching in MUMs limits effectiveness with increased genome divergence and dataset size.
• Existing methods to improve MUM robustness involve trade-offs in specificity, scalability, or computational complexity.

Purpose of the Study:

• To introduce a novel computational approach to enhance the robustness of MUM-based seeding for genome alignment.
• To address the limitations of exact string matching in MUM identification for divergent genomes.
• To improve the recovery of homologous regions and increase base pair coverage in multiple genome alignments.

Main Methods:

• The Min-Frame Transformation (MFT) encodes nucleotide sequences into a transformed alphabet, preserving coordinate structure.
• MFT selects k-mers from local windows and maps them to characters, capturing local sequence context and masking mutations.
• Transformed sequences are indexed using standard data structures (suffix arrays, suffix trees) for efficient MUM extraction.

Main Results:

• MFT produces longer and more contiguous homologous matches, increasing alignment coverage.
• The method improves the detection of homologous regions in divergent genomes.
• MFT enhances SNP recall and maintains efficiency without modifying existing alignment algorithms.

Conclusions:

• MFT offers a robust computational approach for improving MUM seeding in genome alignment.
• The transformation enhances the ability to identify homologous regions across divergent genomes.
• MFT has potential applications in improving viral genome analysis, including phylogenetic inference and transmission studies.