A Novel Methodology for a Comprehensive Analysis of Genomic Sequence-to-Graph Alignment Tools

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Genome graphs provide a compact representation of genetic variability, surpassing traditional references.
• Sequence-to-graph alignment is crucial for graph-based genome analysis but computationally intensive.
• Standardized comparison of sequence-to-graph alignment tools is lacking due to complexity and novelty.

Purpose of the Study:

• To propose a methodology for comprehensive comparison of sequence-to-graph alignment tools.
• To introduce the first open-source benchmark suite for quantitative analysis of these aligners.
• To provide guidelines for improving sequence-to-graph alignment algorithms.

Main Methods:

• Defined Key Performance Indicators (KPIs) for qualitative analysis (usability, accuracy, performance).
• Developed an open-source benchmark suite for quantitative assessment of multiple aligners.
• Tested the methodology on state-of-the-art sequence-to-graph alignment tools.

Main Results:

• The proposed methodology facilitates structured comparison of alignment tools.
• The benchmark suite provides valuable insights into aligner performance.
• Demonstrated the effectiveness of the approach on current state-of-the-art tools.

Conclusions:

• The developed methodology and benchmark suite enable effective evaluation of sequence-to-graph aligners.
• This work addresses the need for standardization in comparing genome graph alignment tools.
• Guidelines are provided to advance the field of graph-based genome analysis.