Related Experiment Videos
Optimizing taxonomic classification of marker-gene amplicon sequences with QIIME 2's q2-feature-classifier plugin
Nicholas A Bokulich1, Benjamin D Kaehler2, Jai Ram Rideout3
1The Pathogen and Microbiome Institute, Northern Arizona University, PO Box 4073, Flagstaff, AZ, 86011-4073, USA. nicholas.bokulich@nau.edu.
Microbiome
|May 19, 2018
Summary
QIIME 2’s q2-feature-classifier plugin offers new machine-learning and alignment-based methods for microbiome taxonomic classification. These QIIME 2 classifiers achieve high species-level accuracy for 16S rRNA and ITS sequences, outperforming previous methods.
Area of Science:
- Microbiology
- Bioinformatics
- Computational Biology
Background:
- Accurate taxonomic classification of marker-gene sequences is crucial for microbiome analysis.
- Existing methods in QIIME 1 have limitations in classification accuracy.
Purpose of the Study:
- To introduce and evaluate novel machine-learning and alignment-based taxonomic classifiers within the QIIME 2 framework.
- To compare the performance of new QIIME 2 classifiers against established methods for marker-gene sequence data.
Main Methods:
- Developed and implemented q2-feature-classifier, a QIIME 2 plugin with naive Bayes and alignment-based classifiers (VSEARCH, BLAST+).
- Evaluated classifiers using bacterial 16S rRNA and fungal ITS marker-gene amplicon sequences.
- Utilized 19 mock communities and simulated sequences for benchmarking within the tax-credit framework.
Main Results:
- QIIME 2's naive Bayes, BLAST+-based, and VSEARCH-based classifiers demonstrate comparable or superior species-level accuracy.
- Performance evaluations included error-free simulations and classification of novel marker-gene sequences.
- Parameter optimization is critical for maximizing classifier performance.
Conclusions:
- The q2-feature-classifier plugin provides accurate and reliable taxonomic classification for microbiome studies.
- Recommendations for parameter tuning are provided for standard operating conditions.
- Both q2-feature-classifier and tax-credit are open-source and freely available.