Related Experiment Video
Updated: Apr 19, 2026
07:37
CAPRRESI: Chimera Assembly by Plasmid Recovery and Restriction Enzyme Site Insertion
Published on: June 25, 2017
12.2K
CATCh, an ensemble classifier for chimera detection in 16S rRNA sequencing studies
Mohamed Mysara1, Yvan Saeys2, Natalie Leys3
1Unit of Microbiology, Belgian Nuclear Research Centre (SCK•CEN), Mol, Belgium Department of Bioscience Engineering, Vrije Universiteit Brussel, Brussels, Belgium VIB Center for the Biology of Disease, Leuven, Belgium.
Applied and Environmental Microbiology
|December 21, 2014
Summary
This study introduces CATCh, a novel machine learning method for detecting artificial chimeric sequences in microbial diversity studies using 16S rRNA gene sequencing. CATCh improves accuracy by combining existing tools, leading to more reliable ecological data.
Area of Science:
- Microbial Ecology
- Bioinformatics
- Genomics
Background:
- Microbial diversity assessment commonly uses 16S rRNA gene sequencing.
- PCR amplification introduces artificial chimeric sequences, complicating diversity analysis.
- Existing chimera detection tools have limitations, especially when combining methodologies.
Purpose of the Study:
- To develop a more powerful chimera detection method by integrating existing tools.
- To improve the accuracy and robustness of chimera removal in microbial sequencing data.
- To enhance the quality of downstream analyses like operational taxonomic unit clustering.
Main Methods:
- Developed two machine learning classifiers: reference-based and de novo CATCh.
- Integrated outputs from multiple existing chimera detection algorithms.
- Evaluated performance on simulated, 454 pyrosequencing, and Illumina MiSeq datasets.
Main Results:
- The ensemble CATCh method demonstrated higher performance compared to individual tools.
- Achieved robust results with challenging chimeric sequences (low parent divergence, short length, multiple parents).
- Integration of CATCh improved the quality of operational taxonomic unit clustering.
Conclusions:
- CATCh offers a superior ensemble approach for chimera detection in microbial ecology.
- This method enhances the reliability of microbial diversity assessments.
- Improved chimera removal leads to more accurate ecological and genomic data interpretation.
