miniasm

miniasm constructs de novo genome assemblies from noisy long reads produced by Single Molecule Real-Time (SMRT) sequencing and Oxford Nanopore Technologies (ONT) using an overlap-layout approach without prior error correction.

Key Features:

• Overlap-Layout-Consensus (OLC): Implements an overlap-layout-consensus (OLC) approach for de novo assembly without requiring a prior error correction stage.
• Input from minimap: Consumes all-vs-all read self-mappings generated by the mapper minimap as input.
• Assembly graph output: Constructs an assembly graph in Graphical Fragment Assembly (GFA) format.
• Long-read support: Targets long reads exceeding 10 kilobases from SMRT and ONT with inherent error rates up to 10–15%.
• Speed: Enables rapid assembly that significantly outperforms traditional pipelines in speed.
• Pairwise mapping format: Introduces a novel pairwise read mapping format for representing overlaps.
• Interoperability: Supports interoperability with other bioinformatics tools through use of the GFA format.

Scientific Applications:

• De novo genome assembly: Performs rapid de novo assembly of genomes from noisy long-read sequencing data (SMRT, ONT).
• Workflow integration: Integrates into broader genomic analysis workflows via GFA output and compatibility with minimap mappings.

Methodology:

Applies an overlap-layout-consensus (OLC) approach to all-vs-all minimap read self-mappings, constructs an assembly graph in GFA format, and uses a novel pairwise read mapping format while omitting a prior error-correction stage.