Experimental Metagenome on MinION

Experimental Metagenome on MinION performs long-read nanopore sequencing and metagenomic analysis to assign taxonomy and characterize microbial community composition and function in environmental samples.

Key Features:

• Long-Read Sequencing: Generates bidirectional high-quality (2D) reads with average lengths of approximately 5,700–6,000 bp and can produce reads ≥10,000 bp to support long-read-based taxonomic and functional analyses.
• Assignment Accuracy: Achieves genus-level assignment accuracy of 53.1%–99.5% and species-level accuracy of 23.9%–99.5%, with misassignments primarily occurring among closely related organisms.
• Synthetic Metagenome Performance: In low-complexity synthetic communities obtained up to 98% species-level assignment accuracy, and in a 20-strain community detected all but three species with 91% correct species assignments and >99% correct family assignments.
• Community Proportion Detection: Reflects known community compositions with deviations ranging from 0.1% to 10%.
• Bioinformatic Integration: Long reads can be analyzed with Kraken and One Codex to enable rapid taxonomic assignment in communities comprising a limited number of taxa.
• Sequencing Platform and Consumables: Sequencing was performed on the original MinION device using R7.3 flow cells and library kits versions 5 and 6.
• Sensitivity and Throughput Limitations: Performance is sensitive to input DNA quality, varies across libraries and flow cells, and limited read counts can hinder detection of very rare taxa in high-complexity metagenomes.

Scientific Applications:

• Environmental metagenomics: Precise taxonomic assignment and functional characterization of microbial communities in environmental samples.
• Defined consortia and controlled experiments: Analysis of low-complexity or defined microbial consortia and controlled experimental setups.
• Real-time community monitoring: Potential for real-time analysis of microbial community composition and structure, contingent on improvements in sequence throughput and error rates.

Methodology:

Long reads are analyzed with bioinformatics tools such as Kraken and One Codex to assign taxonomy at genus and species levels.