Metabarcoding - Bioinformatics Institute

Report
Metabarcoding
16S RNA targeted sequencing
Peter Tsai
Bioinformatics Institute, University of Auckland
Overview


What’s metagenomics and metabarcoding?
Next generation sequencing and
metabarcoding
 How NGS changes metagenomics

Analysis approach
 Taxonomic dependent and independent analysis

Study example
 NZ vine yeast biogeography by pyrosequencing
Metagenomics
Metagenomics
 Study of metagenomes, genetic materials directly from
environmental samples.
 Shotgun metagenomics
◦ Randomly shears DNA, sequence many different species in environment
and attempts to reconstruct multiple genomes.
Metabarcoding
 Subset of metagenomics.
 Study of one or more marker gene.
 Gene specific primers to ‘barcode’ that gene, i.e. 16S, ITS or CO1
 Aim is often to identify different species and compare different
community
NGS and metagenomics




Accelerated by NGS, predominately 454 sequencing
because of the longer read length, now more with Illumina
based chemistry.
Organism no longer needs to be cultivated and cloned —
Culture independent insight
Direct sequencing from environment as a “community”
You can pool multiple samples together
Not all microbes can be cultured
Analysis approach
Analysis approach

Taxonomy independent analysis
 Reads are group into operational taxonomic units (OTU)
based on a specified sequence variation.

Taxonomy dependent analysis
 Assignment at the level of domain, phylum, class, order,
family, genus, and species
 Require a reference database
Taxonomy independent analysis

Group reads into OTU based on certain imposed
similarity threshold
 In study of bacteria, 97% seems like a good starting point
 Species dependent, genes dependent, threshold may vary
 1 OTU = 1 organism

Extract a OTU representative sequence
 Most common sequence
 Sequence that has minimum difference to all other
sequences in the same OTU
Taxonomy dependent analysis


Classify sequences
BLAST
 Simply BLAST what you have

Online RDP classifier (Ribosomal Database Project )
 RDP 10.26 (Release 10, Update 26 consists of 1,613,063 aligned and
annotated 16S rRNA sequences
 Limited by number of reads you can submit

Online Greengenes classifier based on NAST alignment
 Require pre-aligned dataset
 Limited by number of reads you can submit
NZ vine yeast biogeography
by pyrosequencing
M. W. Taylor, N. Anfang, A. H. Thrimawithana, P. Tsai, H. Ross and M. R. Goddard
School of Biological Sciences, University of Auckland
NZ vine yeast biogeography by pyrosequencing




Yeasts are the agents responsible for fermentation of
fruits into wine
Yeasts naturally associated with vines and wines are
reasonably well characterised
Microbes have an effect on both vine and fruit
development (as some are pathogens), as well as the
resulting wine quality and style
Investigations into the ecology of these organisms is
lacking.
Vitis vinifera
NZ vine yeast biogeography by pyrosequencing

6 distinct vineyards in each of four major and distinct
wine-producing regions





West Auckland (WA)
Hawke’s Bay (HB)
Marlborough (MB)
Central Otago (CO)
26S RNA gene from DNA directly extracted from
microbial communities associated with ripe
Chardonnay fruit
NZ vine yeast biogeography by pyrosequencing

Quality checks
◦ Remove short reads
◦ Remove reads containing ambiguity
◦ Trim off low quality regions

Taxonomy independent analysis
◦
◦
◦
◦
No well established reference database for eukaryotic 26S
Clustering into 98% OTU
ANOSIM for statistical test between regions
Limited classification rely upon NCBI Taxonomy DB
NZ vine yeast biogeography by pyrosequencing




2,000 species were found using deep sequencing across all regions.
Culture based analysis recovered 7 species from West Auckland and
Hawke’s Bay
Deep sequencing identified ~700 from the same West Auckland and
Hawke’s Bay sample.
All 7 species were found in pyrosequencing dataset
The culture-based may miss ~99% of the community
Marlborough
Hawke’s Bay
Central Otago
West Auckland
Geographic patterns for yeast communities

Central Otago harbours the most distinct community

Different communities associated with Chardonnay vines in
different areas of NZ

Community similarity significantly decays with distance and
temperature

Different regions harbour different communities, may, in part,
contribute to the distinctiveness of wines deriving from that
area.
Key questions associated with Metagenomics

Number of reads needed
 Statistical power

Over estimating due to sequencing error
 Results in large number of OTUs

Multiple copies of 16S rRNA gene in some species
 Lead to overrepresentation


Accuracy of taxonomic classification
Not all rRNA genes amplify equally well with the same
“universal” primers
Summary

Basic introduction, basic method, one of many ways of
analysing metabarcoded dataset.

Increasingly popular way of extracting the genomes of microorganisms.

Direct insight into communities without the need of culturing

Culture based and sequencing based method may recover
different proportion of organisms

similar documents