metaMIC

metaMIC identifies and corrects misassemblies in metagenomic assemblies to improve the accuracy of metagenome-assembled genomes (MAGs) and downstream analyses.

Key Features:

• Machine-learning misassembly detection: Uses a random forest classifier to detect misassembled contigs in metagenomic assemblies.
• Feature extraction from alignments: Extracts features from the alignment between paired-end sequencing reads and assembled contigs.
• Breakpoint localization and correction: Localizes breakpoints in misassembled contigs and corrects errors by splitting contigs at identified breakpoints.
• Downstream improvement: Correction of misassemblies enhances scaffolding and binning and supports construction of reliable metagenome-assembled genomes (MAGs).
• Benchmarking: Evaluated on simulated and real datasets and reported to outperform existing tools.

Scientific Applications:

• MAG reconstruction: Improvement and validation of metagenome-assembled genomes (MAGs) from metagenomic assemblies.
• Scaffolding and binning enhancement: Reduction of assembly errors to increase accuracy of scaffolding and binning workflows.
• Microbial community analysis: Enabling more accurate downstream analyses of microbial communities and their functions.

Methodology:

MetaMIC extracts features from alignments between paired-end sequencing reads and assembled contigs; applies a random forest classifier to identify misassembled contigs; localizes breakpoints and corrects misassemblies by splitting contigs at identified breakpoints.