Success of DNA barcoding in distinguishing sister species of ...
Transcript of Success of DNA barcoding in distinguishing sister species of ...
Success of DNA barcoding in distinguishing sister species of diverse
clades of birds
Allan Baker, Erika Tavares, Rebecca Elbourne
Department of Natural History, Royal Ontario Museum
Organization of the talk
(1) Sister-species comparisons need to be tested
(2)10X rule for the ratio of intra- to interspecific distance fails to detect sister and cryptic species
(3) Multiple genes are needed to barcode species uniquely
(4) More extensive sampling of intra-specific diversity
(5) Problems with hybridization and paralogy in birds
Aim is to address the following concerns about DNA barcoding:
Barcoding sister-species of birds
Multigene Multigene phylogeniesphylogenies
Sister-species Sister-species selectionselection
QuickTime™ and aTIFF (Uncompressed) decompressor
are needed to see this picture.
COI alignmentCOI alignment
DNA DNA extractionextraction
Amplification and Amplification and sequencing of COIsequencing of COI
GenBank GenBank and BOLD and BOLD
searchsearch
Calculate COI Calculate COI corrected corrected
distances distances
Map COI of Map COI of sister-sister-
species in species in multigene multigene phylogenyphylogeny
Multigene phylogeny of terns
* Posterior Prob. = 1.00
Bayesian analysis 12S, Cytb, ND2, COI :3320 bp Bridge et al 2005, MPE 35: 459-469
COI:651 bpDistances
Multigene phylogeny of shanks Bayesian analysis12S and 16S rDNAs, ATPase8, ATPase6, COIII, ND1, tRNA-ISO, ND2, ND5, Cytb: 7664 bp. Pereira and Baker 2005 Condor 107.
COI:651 bpDistances
* Posterior Prob. = 1.00
Multigene phylogeny of Calidridine sandpipers COI:~763 bpDistances
Bayesian analysisRAG1 + mtDNA: 12,143 bp. Baker et al. In prep.
Multigene phylogeny of penguins COI:~615 bpDistances Bayesian analysis
RAG1, 12S, 16S, COI, Cytb: 5571 bp.
Pereira and Baker et al (2006) Proc. Biol. Sci
273.
* Posterior Prob. = 1.00
Multigene phylogeny of kiwis
COI:~650 bpDistances
Maximum LikelihoodATPAse 6, ATPase 8, CR, Cytb: 2267 bp.
Burbidge et al (2003) Conservation
Genetics 4.
* Bootstrap proportion = 100%
Multigene phylogeny of Pteroglossus toucans COI:~615 bpDistances
Bayesian analysisATPase, COI, Cytb: 2168 bp.
Eberhard and Bermingham (2005) MPE: 288-304
* Posterior Prob. = 1.00
Multigene phylogeny of Woodpeckers
Bayesian analysisCOI, Cytb: 2580 bp.
Moore et al (2006) Biol. J. Linn. Soc.87
* Posterior Prob. = 1.00
COI:~660 bpDistances
Sister-species barcode comparisons
Do many genes improve on COI barcoding of sister-species in 7 clades of birds?
Multigene Multigene phylogeniesphylogenies
Sister-species Sister-species selectionselection
QuickTime™ and aTIFF (Uncompressed) decompressor
are needed to see this picture.
COI alignment of COI alignment of well sampled well sampled species species DNA DNA
extractionextraction Amplification and Amplification and sequencing of COIsequencing of COI
GenBank GenBank and BOLD and BOLD
searchsearch
Calculate COI Calculate COI distances and NJ distances and NJ topologies: K2Ptopologies: K2P
Test barcode in
COI vs multigenes
Success in barcoding sister-species of birds
Sample of 812 sequences from 95 sister-species pairs
2 Clusters 1 Cluster
COI 93.7% (89) 6.3%(6)
Multigene 93.7% (89) 6.3%(6)
Of the 6 “failures” 4 were known to be from hybridization
Conclusion: No improvement using more genes
Intraspecific versus among sister-species variation
Sample of 812 sequences from 95 sister-species pairs.
All Birds
0
10
20
30
40
50
60
70
80
0.5 1 2 3 4 5 6 7 8 9 10
Percent Divergence (Kimura 2P)
Fre
qu
en
cy
Intraspecific
Sister-species
Conclusion: This is a potential problem for barcoding sister-species?
Within-species variation in terns
Terns
Intra-specific
Between sister-species
Within genera (above sister species level) Between genera
Neighbour Joining (Kimura-2P) Node support: Posterior Prob. Bayesian analysis.
COI (651 bp)COI (651 bp)Multigene 3220 bp)Multigene 3220 bp)
New species of Little terns?
1.5%
1.3%
http://www.birdguides.com
Large-bill form
Small-bill form
1.7%
1.8%
New species of Gull-billed terns?
Gull-billed Tern
Within-species variation in Tringid sandpipers
Intra-specific
Between sister-species
Within genera (above sister species level) Between genera
Neighbour Joining (model Kimura-2P) Node support: Posterior Prob. Bayesian analysis.
COI (623 bp)COI (623 bp)Multigene (7664 bp)Multigene (7664 bp)
*
Intra-specific
Between sister-species
Within genera (above sister species level) Between genera
Within-species variation in penguins
PenguinsCOI (bp)COI (bp)Multigene (bp)Multigene (bp)
Neighbour Joining (model Kimura-2P)Node support: Posterior Prob. Bayesian analysis.
2- White-Flippered PenguinEudyptula m. albosignata
(J.N. Davis and J.Busby)
3-Eudyptula m. novaehollandiae
(J.N. Davis and J.Busby)
1-Little PenguinEudyptula m. minor
(J.N. Davis and J.Busby)
a,b
a,b
c,d
a,b
e,f
g,h
Banks , JC. 2002 Notornis 49:29-38.
Northern a
Northern b
Banks Peninsula b
Banks Peninsula a
Cook Strait c
Cook Strait d
Chatham Is. e
Chatham Is. f
Otago g
Otago h
Australia a
Australia b
COI barcode
3.56%
Total Evidence Tree
New species of Little blue penguins
1%
New species of kiwis
COI 650 bpCOI 650 bpMultigene phylogenyMultigene phylogeny
Threshold for sister-species
Terns
0
5
10
15
20
25
0.5 1 2 3 4 5 6 7 8 9 10 15 20 30 40 50 60 70 80 90 100
Distances (TN+I+G)
Freq
uenc
y
Penguins
0
2
4
6
8
10
12
14
16
18
0.5 1 2 3 4 5 6 7 8 9 10 15 20 30 40 50 60 70 80 90 100
Distances GTR+I+G
Freq
uenc
y
Tringid sandpipers
0
2
4
6
8
10
12
0.5 1 2 3 4 5 6 7 8 9 10 15 20 30 40 50 60 70 80 90 100
Distances (GTR+I+G)
Fre
qu
ency
sister species
intraspecific
Toucans
0
2
4
6
8
10
12
0.5 1 2 3 4 5 6 7 8 9 10 15 20 30 40 50 60 70 80 90 100
Distances (GTR+I+G)
Fre
qu
ency
sister species
intraspecific
Excluding putative new species and hybridizing species
Calidridine sandpipers
0
2
4
6
8
10
12
14
16
0.5 1 2 3 4 5 6 8 9 10 11
Percent Divergence
Fre
qu
en
cy
Intraspecific
Sister-species
Threshold for sister-species
0
2
4
6
8
10
12
14
16
0.5 1 2 3 4 5 6 7 8 9 10 15 20 30 40 50 60 70 80 90 100
Fre
qu
ency
Distance (GTR+I+G)
Woodpeckers
intraspecif ic
sister-species
Conclusion: Sister-species and intraspecific variation overlap slightly around 1% level
Threshold for sister-species
COI barcodes vs time of divergence of species
Estimation of Estimation of divergence divergence times if not times if not available available
with with multidivtimemultidivtime
Multigene Multigene phylogeniesphylogenies
QuickTime™ and aTIFF (Uncompressed) decompressor
are needed to see this picture.
COI alignmentCOI alignment
COI distances vs COI distances vs divergence timesdivergence times
distances x time
0
2
4
6
8
10
12
14
16
18
0 5 10 15 20 25
divergence times
dis
tan
ce
s T
N+
g+
I
Model choiceModel choice
Species barcode level varies with time since speciation
0
2
4
6
8
10
12
14
16
18
20
0 5 10 15 20 25
Div
erg
ence
Tim
es (
MY
)
Distances (GTR+I+G)
Sister-species Distances
Penguins
Curassous
Terns
Shanks
Summary for these clades
• The levels of distances between sister-species pairs varies from 1% to more
than 20%.
• When putative new species and hybridizing forms are excluded, COI
distances (>1.3% ) between sister-species do not overlap with intraspecific
variation
• However, all sister-species have unique DNA barcodes that identify them
• Results are specific to the clades we investigated, and potentially will vary with
different rates of evolution in birds with faster rates of evolution
• These phenomena can be flagged with COI barcoding
• By exposing these difficult cases barcoding will speed higher resolution
studies required to make taxonomic decisions on species limits
• Most species lose the ability to hybridize within 1-2 million years after
divergence, so it should not be a pervasive problem in the tree of life
• Even in closely related sister-species of birds less than 6% hybridize, so
barcoding is highly effective in cataloguing species diversity and discovering
putative new species.
Hybridization, unsorted polymorphism and paralogy
Acknowledgements
• This research was supported by funding to the Canadian Barcode of Life
Network from Genome Canada through the Ontario Genomics Institute,
NSERC, and other sponsors listed at www.BOLNET.ca.
• We thank Oliver Haddrath, Sergio Pereira, Eli Bridge and the Hebert lab for
tips and assistance, and Rob Brumfield,Joel Cracraft and Bob Zink for
permission to barcode loaned samples.