DBH: A de Bruijn graph-based heuristic method for clustering large-scale 16S rRNA sequences into OTUs

Area of Science:

• Microbiology
• Bioinformatics
• Computational Biology

Background:

• High-throughput sequencing generates vast microbial 16S rRNA data.
• Clustering sequences into operational taxonomic units (OTUs) is essential for metagenomic analysis.
• Existing methods have limitations in seed selection and sensitivity.

Purpose of the Study:

• To develop a novel heuristic clustering method for large-scale 16S rRNA sequence analysis.
• To improve the accuracy and efficiency of operational taxonomic unit (OTU) inference.
• To address the limitations of seed selection in current OTU clustering algorithms.

Main Methods:

• A de Bruijn graph-based heuristic clustering approach (DBH).
• Introduction of a novel seed selection strategy.
• Application of a greedy clustering algorithm.

Main Results:

• DBH demonstrates higher clustering performance compared to existing methods.
• DBH exhibits low memory usage, suitable for large datasets.
• The method effectively handles massive 16S rRNA sequence data.

Conclusions:

• DBH provides a more effective and accurate solution for OTU clustering in metagenomics.
• The developed method is efficient for analyzing large-scale microbial community data.
• DBH software is available for academic use.