Merqury: reference-free quality, completeness, and phasing assessment for genome assemblies.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Long-read sequencing technologies enable the generation of high-quality genome assemblies.
• Validating these advanced assemblies against existing reference genomes is often difficult due to superior quality and completeness.
• There is a need for robust, reference-free methods to evaluate novel genome assemblies.

Purpose of the Study:

• To introduce Merqury, a novel computational tool for reference-free genome assembly evaluation.
• To assess the accuracy, completeness, and haplotype-specific metrics of genome assemblies.
• To provide efficient and versatile visualization tools for assembly quality assessment.

Main Methods:

• Merqury utilizes efficient k-mer set operations for comparative analysis.
• It compares k-mers from a de novo assembly against k-mers from unassembled, high-accuracy sequencing reads.
• For trio data, Merqury extends analysis to haplotype-specific accuracy, completeness, and phasing quality.

Main Results:

• Merqury accurately estimates base-level accuracy and completeness of genome assemblies.
• The tool successfully evaluates haplotype-specific accuracy, completeness, and phase block continuity in trios.
• K-mer spectrum plots and other visualizations aid in comprehensive assembly evaluation.
• Demonstrated efficiency and robustness on both human and plant genome assemblies.

Conclusions:

• Merqury provides a fast, accurate, and robust method for reference-free genome assembly validation.
• The tool is applicable to diverse genomic datasets, including human and plant genomes.
• Merqury addresses the challenge of evaluating high-quality assemblies in the absence of a reference.