NextPolish

NextPolish polishes genome assemblies by correcting sequencing errors in long-read assemblies using short reads, long reads, or both to improve base-level accuracy.

Key Features:

• Dual-Module Design: Two interlinked modules—one scores and counts k-mers derived from high-quality short reads, and the other polishes the assembly to correct base errors such as single nucleotide variants (SNVs) and insertions/deletions (Indels).
• Versatile Data Inputs: Supports polishing using short read data alone, long read data alone, or a combination of short and long reads.
• Error Types Corrected: Targets base-level errors including SNVs and Indels in assemblies generated from long-read sequencing technologies.
• Benchmark Performance: Demonstrated higher speed and improved correction accuracy compared with Pilon on human and Arabidopsis thaliana assemblies.
• Implementation: Implemented in C and Python.

Scientific Applications:

• Genome assembly polishing: Enhances base-level accuracy of assemblies produced from noisy long-read sequencing technologies.
• Variant calling: Improves the accuracy of downstream variant calling by reducing assembly-derived errors.
• Comparative genomics: Raises assembly quality for comparative analyses across species or strains, including human and Arabidopsis thaliana.
• Functional annotation: Reduces false-positive and false-negative signals in gene models and other annotation tasks by correcting sequence errors.

Methodology:

Operates with two modules: the first scores and counts k-mers from short reads, and the second polishes assemblies to correct SNVs and Indels; supports using short reads, long reads, or both for polishing.