Online Program Home
  My Program

Abstract Details

Activity Number: 445 - Quantification, Association Testing, and Integration of Micriobiome
Type: Invited
Date/Time: Wednesday, August 2, 2017 : 8:30 AM to 10:20 AM
Sponsor: WNAR
Abstract #322002 View Presentation
Title: BetterĀ Measurement of the Microbiome by Modeling and RemovingĀ Amplified Sequencing Errors
Author(s): Benjamin John Callahan*
Companies: North Carolina State University
Keywords: microbiome ; bioinformatics ; models ; sequencing
Abstract:

PCR amplification and high-throughput sequencing (amplicon sequencing) of phylogenetically informative genetic loci such as the 16S rRNA gene is the most common method used to characterize complex microbiomes. Amplicon sequencing introduces errors of multiple types - including substitution errors, chimeric DNA sequences, and contaminating DNA - that obscure the true composition of the sampled community. Amplicon errors have traditionally been discriminated from biological variation with ad hoc thresholds, such as the lumping of sequences within a fixed edit-distance into operational taxonomic units (OTUs). Quantitative models of the amplicon error process can better discriminate between biological and artifactual sequences, thereby improving the accuracy and resolution of amplicon sequencing. The DADA2 algorithm uses a quantitative model of the substitution process to correct point-errors in amplicon sequences, and outperforms OTU methods in both sensitivity and specificity. Similarly, a model of relative abundance across samples can be used to remove contaminating sequences, extending the effective sensitivity of amplicon sequencing to rare variants.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association