Multi-amplicon microbiome data analysis pipelines for mixed orientation sequences using QIIME2: Assessing reference...

Area of Science:

• Microbiology
• Bioinformatics
• Genomics

Background:

• Microbiome research utilizes next-generation sequencing and complex data analysis.
• Multi-amplicon kits enhance microbiome speciation but introduce analytical challenges.
• Mixed-orientation reads from multi-hypervariable regions require specialized bioinformatics approaches.

Purpose of the Study:

• To develop and present two distinct analysis pipelines for mixed-orientation reads from multi-hypervariable (V) region amplicons.
• To assess the agreement of sequencing data with expected abundances, considering different variable regions and reference databases.
• To provide a benchmark for microbiome researchers using multi-amplicon sequencing kits.

Main Methods:

• Two analysis workflows were developed: one using a specialized plugin (CutPrimers) and another using Cutadapt.
• Mock community sequence data from the Ion16S™ Metagenomics Kit on the Ion Torrent Platform were analyzed.
• Taxonomic assignment was performed using three reference databases: Ribosomal Database Project, Greengenes, and Silva.
• Annotation consistency was evaluated using Bray-Curtis, Euclidean, and Jensen-Shannon distances.

Main Results:

• Both CutPrimers and Cutadapt-based methods showed variability in read mapping across V2-V9 regions.
• The CutPrimers pipeline demonstrated the best agreement with expected amplicon distribution for V3 regions, while V9 regions showed the poorest agreement.
• Accurate taxonomic annotation varied significantly depending on the genus-level taxon and the specific V region analyzed.
• Overall agreement with expected mock community abundances differed across variable regions and reference databases.

Conclusions:

• The study presents novel microbiome analysis pipelines for separating multi-amplicon data into V-specific reads, specifically for the Ion16S Metagenomics Kit and Ion Torrent platform.
• The findings highlight significant biases introduced by variable regions and reference databases in microbiome data analysis.
• Researchers should consider this benchmarking data when planning multi-amplicon sequencing studies to mitigate potential biases.