Talk given at UNC Charlotte Bioinformatics department about Open Tree of LIfe

1. OPEN TREE OF LIFE:
SYNTHESIZING
PHYLOGENETIC DATA
Karen Cranston
National Evolutionary Synthesis Center (NESCent)
@kcranstn
http://www.slideshare.net/kcranstn

2. Tree of life
• ~2million named
species
• Millions
more
unnamed / undiscovered

3. is that golden moles, a family of insectivores, are also part of this
clade. 12S rRNA sequences earlier suggested an association of
golden moles with paenungulates, but did not provide convincing
bootstrap support for this hypothesis7. Our expanded data set
Endemic African mammals demonstrates that insectivores are not monophyletic (Table 2)
shake the phylogenetic tree
Table 1 Bootstrap support for select clades based on different methods
Mark S. Springer*, Gregory C. Cleven*, Ole Madsen†,
Wilfried W. de Jong†‡, Victor G. Waddell§, Clade
Heather M. Amrine* & Michael J. Stanhope§ Paenungulata Paenungulata aardvark
elephant shrew
* Department of Biology, University of California, Riverside, California 92521, golden mole
USA Mitochondrial DNA
† Department of Biochemistry, University of Nijmegen, PO Box 9101, Parsimony 99 95
6500 HB Nijmegen, The Netherlands Transversion parsimony 64 90
Minimum evolution
‡ Institute for Systematics and Population Biology, University of Amsterdam, PO
Tamura–Nei I 100 92
Box 94766, 1090GT Amsterdam, The Netherlands Tamura–Nei II 100 78
§ Biology and Biochemistry, Queen’s University, 97 Lisburn Road, Logdet 99 90
Belfast BT9 07BL, UK Maximum likelihood 100 100
......................................................................................................................... vWF
Parsimony
The order Insectivora, including living taxa (lipotyphlans) and All positions 49 99
archaic fossil forms, is central to the question of higher-level 1st and 2nd positions 24 65
relationships among placental mammals1. Beginning with 3rd positions 51 93
Transversion parsimony 30 95
Huxley2, it has been argued that insectivores retain many primi- Minimum evolution
tive features and are closer to the ancestral stock of mammals than Tamura–Nei I 37 99
are other living groups3. Nevertheless, cladistic analysis suggests Tamura–Nei II 30 99
Logdet 43 97
that living insectivores, at least, are united by derived anatomical Maximum likelihood 78 100
features4. Here we analyse DNA sequences from three mito- A2AB
chondrial genes and two nuclear (2001) to examine relationships
Hedges, B. PNAS 98 genes Parsimony
All sites 71 88
of insectivores to other mammals. The representative insectivores 1st and 2nd positions 49 81
are not monophyletic in any of our analyses. Rather, golden moles 3rd positions 31 67
are included in a clade that contains hyraxes, manatees, elephants, Transversion parsimony 71 54
Minimum evolution
elephant shrews and aardvarks. Members of this group are of Tamura–Nei I 83 84
presumed African origin5,6. This implies that there was an exten- Tamura–Nei II 28 25
sive African radiation from a single common ancestor that gave Logdet 79 78
Maximum likelihood 81 89
rise to ecologically divergent adaptive types. 12S ribosomal RNA .............................................................................................................................................................................
transversions suggest that the base of this radiation occurred Springer, M.S. et al. Nature (1997)
Only two of the three paenungulate orders were represented among the mitochondrial and
A2AB sequences. Tamura–Nei27 I and II distances were calculated by using an equal-rates
during Africa’s window of isolation in the Cretaceous period assumption and aM.J. et al. PNAS of rates, respectively.
Stanhope, gamma-distribution (1998)
Nature © Macmillan Publishers Ltd 1997

4. EVOLUTION OF HUMAN PATHOGENS
Smith et al, Nature, 2009

5. MODERN BIOLOGY NEEDS TREES
Conventional Evolutionary
statistics assume: trees provide:
Modified from Garland and Carter, 1994

6. Phylogeny'papers,'1978;2008'
12000"
10000"
Number'of'papers'published'
8000"
Rapid"increase"in"applica?ons"of"
phylogeny,"beginning"in"early"1990s"
6000"
4000"
2000"
0"
1978" 979" 980"1981" 982" 983" 984"1985" 986"1987" 988" 989" 990"1991" 992" 993" 994"1995" 996"1997" 998" 999" 000"2001" 002" 003" 004"2005" 006"2007" 008"
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2
Year'
Source:"ISI"Web"of"Science""
graph from David Hillis

8. What does it mean to “have” the tree of life?

9. Archiving sequence data is a
community norm
~4% of all published
phylogenetic trees
Stoltzfus et al 2012

10. “Publishing a tree”
= picture in a PDF
EVOLUTION
Fig. 1. Combined molecular phylogenetic tree for Diptera. Partitioned ML analysis of combined taxon sets of tier 1 and tier 2 FLYTREE data samples (−lnL =
Weigmann et al. PNAS, 2011
344155.6169) calculated in RAxML. Circles indicate bootstrap support >80% (black/bp = 95–100%, gray/bp = 88–94%, white/bp = 80–88%). Nodes with im-
proved bootstrap values resulting from postanalysis pruning of unstable taxa are marked by stars (black/bp = 95–100%, gray/bp = 88–94%, white/bp = 80–
88%). Colored squares on terminal branches indicate the presence, in at least one species of a family, of ecological traits as shown to lower left. The number
of origins of each trait was estimated with reference to the phylogeny, the distribution of each trait among genera within a family, and the known biology of
the organisms.
thermore, a paraphyletic relationship of phorids and syrphids To test this hypothesis, we used a relatively recent phylogenomic
would support the hypothesis that their shared special mode of marker: small, noncoding, regulatory micro-RNAs (miRNAs).

11. Lander et al. Nature 2001

12. Rod asks: Why do we need a database of trees?

13. assembly
alignment
inference
expertise Fig. 1. Combined molecular phylogenetic tree for Diptera. Partitioned ML analysis of combined taxon sets of tier 1 and tier 2 FLYTREE data samples (−lnL =
344155.6169) calculated in RAxML. Circles indicate bootstrap support >80% (black/bp = 95–100%, gray/bp = 88–94%, white/bp = 80–88%). Nodes with im-
proved bootstrap values resulting from postanalysis pruning of unstable taxa are marked by stars (black/bp = 95–100%, gray/bp = 88–94%, white/bp = 80–
88%). Colored squares on terminal branches indicate the presence, in at least one species of a family, of ecological traits as shown to lower left. The number
of origins of each trait was estimated with reference to the phylogeny, the distribution of each trait among genera within a family, and the known biology of
the organisms.
time thermore, a paraphyletic relationship of phorids and syrphids
would support the hypothesis that their shared special mode of
extraembryonic development (dorsal amnion closure) (26)
evolved in the stem lineage of Cyclorrhapha and preceded the
origin of the schizophoran amnioserosa.
To test this hypothesis, we used a relatively recent phylogenomic
marker: small, noncoding, regulatory micro-RNAs (miRNAs).
miRNAs exhibit a striking phylogenetic pattern of conservation
across the metazoan tree of life, suggesting the accumulation and
maintenance of miRNA families throughout organismal evolution
$$$ Wiegmann et al. PNAS Early Edition | 3 of 6

15. NSF IDEAS LAB
i. Pre-proposal / application iv. Pitch high risk proposal
ideas at end
ii. 5 day highly facilitated
workshop v. NSF invited full proposals
iii. Self-assembly into groups

16. 1. Synthesize a complete draft tree of life from existing
phylogenetic trees
2. Release with:
a. ability to improve tree by uploading new data
b. areas of uncertainty / conflict
c. links to source data and analysis methods
d. utilities to download whole tree and subtrees

17. Graph database holding
thousands of input trees with • filter / weight input trees
millions of nodes
• build synthetic trees
• compare to alternate trees
• input new data sets

18. INPUTS
a AACCGTGAA e a c f b d
b ACACGTTAA
computational
phylogenies c AAACGTTAA
d AGACGTTAA method
e AACCGTTAA
f AAACGTGAA
e a c f b d g h i
taxonomies

19. a b c d
a
b
c
d
a,b c,d
a,b,c,d
a b c d
a
c
b
d
a,b c,d
a,b,c,d
Stephen Smith, U Michigan

20. a b c d
a
c
b
d
a,b c,d
a,b,c,d
a b c d
a,c a,b c,d b,d
a,b,c,d
Stephen Smith, U Michigan

21. a b c d
a
d
e
f
g
h
i
a,c a,b c,d b,d
a,b,c,d
a b c d e f g h
a,c a,b c,d b,d
a,b,c,d e,f,g,h,i
a,b,c,d,e,f,g,h,i
Stephen Smith, U Michigan

22. a b c d e f g h a b c d e f g h
a,c a,b c,d b,d a,c a,b c,d b,d
a,b,c,d e,f,g,h,i a,b,c,d e,f,g,h,i
a,b,c,d,e,f,g,h,i a,b,c,d,e,f,g,h,i
a
c
b
d
e
f
g
h
i
Stephen Smith, U Michigan

23. Dipsicales graph
taxonomy data (578 taxa) +
Soltis et al APG III phylogeny (30 taxa)

24. Dipsicales graph Synthesized tree (favouring
phylogenetic branches); contains
all 578 taxa

25. community
refinement
r evolutionary relationships recovered in this analysis of Carnivora. Illustrations of
ndinia binotata; Felidae (Lynx rufus); Viverridae (Viverra zibetha); Hyaenidae (Crocuta
education and
outreach
ns (Eupleres goudotii); Canidae (Canis lupus); Ursidae (Ursus americanus); Phocidae
ae (Odobenus rosmarus); Ailurus fulgens; Mephitidae (Mephitis mephitis); Procyonidae
d schematic representing diverse taxa [African polecat and striped marten, badger,
stela (Mustela frenata); Mustelidae, Lutrinae (Lontra canadensis).
volutionary relationships recoveredin this analysis of Carnivora. Illustrations of
jor evolutionary relationships recovered in this analysis of Carnivora. Illustrations of
Nandinia binotata; Felidae (Lynxrufus); Viverridae (Viverra zibetha); Hyaenidae (Crocuta
nia binotata; Felidae (Lynx rufus); Viverridae (Viverra zibetha); Hyaenidae (Crocuta
ans (Eupleres goudotii); Canidae (Canis lupus); Ursidae (Ursus americanus); Phocidae
Eupleres goudotii); Canidae (Canis lupus); Ursidae (Ursus americanus); Phocidae
dae (Odobenus rosmarus); Ailurus fulgens; Mephitidae (Mephitis mephitis); Procyonidae
Odobenus rosmarus); Ailurus fulgens; Mephitidae (Mephitis mephitis); badger,
ed schematic representing diverse taxa [African polecat and striped marten, Procyonidae
chematic representing diverse taxa [African polecat and striped marten, badger,
ustela (Mustela frenata); Mustelidae, Lutrinae (Lontra canadensis).
a (Mustela frenata); Mustelidae, Lutrinae (Lontra canadensis).
representing the major evolutionary relationships recovered in this analysis of Carnivora. Illustrations of
nclude (from top): Nandinia binotata; Felidae (Lynx rufus); Viverridae (Viverra zibetha); Hyaenidae (Crocuta
; Malagasy carnivorans (Eupleres goudotii); Canidae (Canis lupus); Ursidae (Ursus americanus); Phocidae
ifornianus); Odobenidae (Odobenus rosmarus); Ailurus fulgens; Mephitidae (Mephitis mephitis); Procyonidae
mustelids (generalized schematic representing diverse taxa [African polecat and striped marten, badger,
gulo); Mustelidae, Mustela (Mustela frenata); Mustelidae, Lutrinae (Lontra canadensis).
?
automatic ?
updating ?
?
downstream
analyses

26. • provide complete phylogenetic
framework
• link to biodiversity and systematics
content
• API for downloading subtrees to analysis tools
• source / storage of underlying data

27. ?
• Open Data
• increasing
availability of digital data associated with
phylogeny publications
• synthetic
tree open to community annotation and
new data submission
• whole tree / subtrees available for download

28. ?
• Open Science
• project wiki: http://opentree.wikispaces.com/
• open source software: https://github.com/OpenTreeOfLife
• public mailing list, meeting notes, management tools

29. Karen Cranston, lead PI (Duke)
Gordon Burleigh (Florida)
Keith Crandall (BYU)
Karl Gude (MSU)
David Hibbett (Clark)
Mark Holder (Kansas)
Laura Katz (Smith)
opentreeoflife.org Rick Ree (FMNH)
Stephen Smith (Michigan)
Doug Soltis (Florida)
Tiffani Williams (TAMU)
AVAToL: Assembling, Visualizing and Analysis of the Tree of Life