metaMDBG

metaMDBG assembles metagenomic long-read sequencing reads, particularly PacBio HiFi, into genome-scale assemblies using a minimizer de Bruijn graph approach to recover high-quality prokaryotic MAGs, plasmids, and viruses.

Key Features:

• Minimizer De Bruijn Graph Assembly: Implements a de Bruijn graph in minimizer space (MDBG) for assembly of long-read metagenomic data.
• Iterative Assembly Over Minimizers: Iteratively assembles sequences of minimizers to manage variations in genome coverage across species.
• Abundance-Based Filtering Strategy: Applies an abundance-based filtering method that prioritizes sequences by relative abundance to simplify strain complexity.
• Multi-k Approach in Minimizer Space: Uses an efficient multi-k strategy that dynamically adjusts k-mer sizes within minimizer space to accommodate varying coverage depths.

Scientific Applications:

• Prokaryotic MAG recovery: Recovers circularized prokaryotic metagenome-assembled genomes (MAGs), reporting up to twice as many high-quality MAGs compared to existing methods.
• Viral and plasmid recovery: Improves recovery rates for viruses and plasmids from metagenomic assemblies.
• Long-read metagenomic assembly: Assembles high-quality metagenomic data from diverse biological samples generated with long-read sequencing technologies such as PacBio HiFi.

Methodology:

Implements a minimizer de Bruijn graph (MDBG) assembly in minimizer space; iteratively assembles sequences of minimizers; applies abundance-based filtering to prioritize sequences by relative abundance; and employs a multi-k strategy that dynamically adjusts k-mer sizes within minimizer space to accommodate variable coverage depths.