Of clades and clans: terms for phylogenetic relationships in unrooted trees.

Letters
Of clades and clans: terms for phylogenetic
relationships in unrooted trees
Mark Wilkinson1, James O. McInerney2, Robert P. Hirt3, Peter G. Foster1 and
T. Martin Embley3
1Department of Zoology, The Natural History Museum, London, SW7 5BD, UK
2Department of Biology, National University of Ireland, Maynooth, Co. Kildare, Ireland
3Division of Biology, The Devonshire Building, University of Newcastle upon Tyne, Newcastle upon Tyne, NE1 7RU, UK
During the late 1980s, the molecular biology community
became aware that the words ‘homology’ and ‘similarity’
werebeingused inmuchof the scientific literatureas though
they were synonyms. This situation was clarified [1] and
their current usage is usually appropriate. Recently, we
have felt a similar need for standard and indeed, new,
terminology for describing relationships in unrooted phylo-
genetic trees in our teaching ofmolecular evolution.Distinct
terms could help teachers to explain, and students to under-
stand, the fundamental concept of phylogenetic relation-
ships and the important differences between unrooted and
rooted trees. Rooted trees have an explicit ancestral node
that is known, whereas unrooted trees have no such node.
Unrooted trees are common in molecular phylogenetics.
However, much of the terminology used to describe
relationships in trees was developed for rooted trees.
The use of terms and phrases such as ‘clade’, ‘monophyletic
group’, ‘more closely related to’ and ‘sister taxa’ to describe
relationships in unrooted trees is problematic, but not
uncommon [2–7].
To say that some genes or taxa are more closely related
to each other than they are to others is to say that they
have a common ancestor that is not an ancestor of any of
the others (i.e. that they share a more recent common
ancestry) [8]. Similarly, to say that a group is monophy-
letic, or that it is a clade, is to say that all of itsmembers are
more closely related to each other than any of them are to
any non-member. To say that two groups are sister taxa is
to say that they are each others closest relatives [9,10].
Thus, all of these concepts can be specified only by rooted
phylogenetic trees.
It is clear that unrooted trees do convey information
about phylogenetic relationships. For example, they tell
us that some sets of taxa cannot be either monophyletic
or sister taxa under any possible rooting. They tell us
that some groups could be clades or could be sister taxa
given one or more rootings of the tree. Given the increase
in the number of unrooted trees being published, we
suggest that it would be useful to have terms that
provide us with unrooted counterparts of ‘clade’, ‘more
closely related’ and ‘sister groups’. Distinct terms would
help prevent misapplication of rooted tree terminology to
unrooted trees and would avoid the concomitant danger
of misinterpretation. The unrooted relationships to be
described are precisely those that could be converted into
the analogous rooted relationship by rooting the tree.
We propose the term ‘clan’ (from theGaelic for family) as
the unrooted analogue of monophyletic group or clade.
There are two complementary clans for every nontrivial
split or bipartition in an unrooted tree. Were the tree to be
rooted, at least one of the two clans defined by a given split
would necessarily be monophyletic. A trivial split in an
unrooted tree [i.e. between one of the leaves (terminal taxa,
OTUs) and all the others] yields a single clan.
We suggest using ‘adjacent group’ as the unrooted
analogue of sister group. Clans or leaves are adjacent
groups if there is some rooting of the tree in which they
are sister groups. In a fully bifurcating rooted tree, a clade
or leaf has a unique sister group; by constrast, in an
unrooted tree, each has up to three adjacent groups
(Figure 1), underlining the potential for selective misap-
plication of rooted tree terminology. Exceptionally, for
clans that include all but one leaf (corresponding to a
trivial split), that single leaf is the only adjacent group
of those clans. To convey the unrooted analogue of the
rooted ‘some taxa are more closely related to each other
than to some other taxa’, we recommend ‘some taxa are
split from some other taxa’.
Figure 1. An unrooted tree with the ‘clan’ including E, F and G encircled. E, F and G
are ‘split from’ A, B, C and D and vice versa. Arrows indicate the seven possible
rootings under which the encircled clan is also a clade. Its three adjacent groups
are A and B (roots 3, 4 and 5), C and D (roots 2, 6 or 7) and A, B, C and D (root 1); if
the encircled clan is a clade, one of the adjacent groups must be its sister taxon.
Corresponding author: Wilkinson, M. (mw@bmnh.org).
Available online 18 January 2007.
114 Update TRENDS in Ecology and Evolution Vol.22 No.3
www.sciencedirect.com

Our prime concern is that the use of the terminology of
rooted trees to describe relationships in unrooted trees
implies an assumption that the root is in a part of the tree
that would make the use of the rooted terms correct.
Injudicious use of rooted terms could lead to incompatible
assumptions regarding the position of the root. In many
unrooted trees, the root might not be known exactly, but
assuming that it occurs in some part of the tree might be
quite reasonable and, therefore, justify judicious use of the
rooted term. Such assumptions are better stated than
implied.
References
1 Reeck, G.R. et al. (1987) ‘Homology’ in proteins and nucleic acids: a
terminology muddle and a way out of it. Cell 50, 667
2 Ben Ali, A. et al. (2001) Phylogenetic relationships among algae based on
complete large-subunit rRNA sequences. Int. J. Syst. Evol.Microbiol. 51,
737–749
3 Bowe, L.M. et al. (2000) Phylogeny of seed plants based on all three
genomic compartments: extant gymnosperms are monophyletic and
Gnetales’ closest relatives are conifers. Proc. Natl. Acad. Sci. U. S. A.
97, 4092–4097
4 Brochier, C. et al. (2004) Archaeal phylogeny based on proteins of the
transcription and translation machineries: tackling the Methanopyrus
kandleri paradox. Genome Biol. 5, R17
5 Cnor, B. et al. (2003) Maximum likelihood on four taxa phylogenetic
trees: analytic solutions. In RECOMB’03 (Vingron, M. et al., eds), p.
7683, ACM Press
6 Hall, B.G. (2004) Phylogenetic Trees Made Easy, Sinauer Associates
7 Xia, X. (1998) The rate heterogeneity of nonsynonymous substitutions
in mammalian mitochondrial genes. Mol. Biol. Evol. 15, 336–
344
8 Wilkinson, M. (1994) Common cladistic information and its consensus
representation: reduced adams and reduced cladistic consensus trees
and profiles. Syst. Biol. 43, 343–368
9 Kitching, I. et al. (1998) Cladistics: The Theory and Practice of
Parsimony Analysis, Oxford University Press
10 Page, R.D.M. and Holmes, E.C. (1998) Molecular Evolution: A
Phylogenetic Approach, Blackwell Science
0169-5347/$ – see front matter ! 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tree.2007.01.002
Book Review
On Lilliputians and Brobdingnagians
Why Size Matters: From Bacteria to Blue Whales by John Tyler Bonner, Princeton University Press, 2006. US$16.95/£9.95 hbk
(176 pages) ISBN 0 691 12850 2
Andre J. Riveros
Center for Insect Science and ARL, Division of Neurobiology, University of Arizona, Tucson, AZ, 85721-0077, USA
Size is one of the most conspicuous
properties of anything occurring in the
living or inanimate world. We see around
us small and large living and non-living
entities; our universe is enormous and
size even characterizes our immaterial
images of gods and thoughts. We obvi-
ously assign those differences in size by
comparison, accordingwith the perception
of our own size. But, are those differences
in size relevant in the natural world? Does size really
matter? These questions are far from trivial and are shared
by scientific disciplines as diverse as ecology, evolutionary
biology, neurobiology and developmental biology, among
others.
In Why Size Matters, John Tyler Bonner categorically
argues that size is a fundamental factor in evolution; in
his own words: ‘No living entity can evolve or develop
without taking size into consideration’. Bonner starts
with a nice historic view of the human perception of
size, from the development of Galileo’s telescope to
observe distant celestial bodies to van Leeuwenhoek’s
microscope to discover the micro world. Today, such tools
have their modern descendants that range from the
Hubbell telescope to the electron microscope, which give
us a new understanding of the micro and macro levels in
our universe. But for Bonner, size is much more than just
a static physical property in those macro and micro
worlds. In fact, the fundamental argument of his book
gives size a central role as a ‘dynamic’ property (a mover
in evolution and not just a by-product of it) that is
strongly tied to fundamental features of living systems,
such as shape, structure and function. He develops his
argument through five rules that are supported by cor-
relates of size: strength, surface, abundance, complexity
and rates of living processes.
The first three rules are strongly connected to physical
laws or constraints. Strength and surface relate to the fact
that certain magnitudes characterizing an object vary with
different powers as size varies. Surface and strength
increase with a square power of the linear dimension,
similar to the surface area of a sphere. By contrast, weight
and volume increase with a cubic power, similar to the
volume and the weight of the sphere just mentioned. How-
ever, abundance is linked to a more obvious physical law:
biggerentities occupymorespace,whichmakes therelation-
ship between size and abundance the only rule exhibiting a
negative correlation. This is exemplified in living forms, for
which being bigger also means greater resource require-
ments and, consequently, lower population densities.
The other two rules, complexity and rate of living
processes, have been of special interest in previous works
[1,2] by Bonner and a considerable portion of this book is
devoted to them. On the one hand, changes in size go along
with increases or decreases in complexity, which, in Bon-
Corresponding author: Riveros, A.J. (ajosafat@email.arizona.edu).
Available online 2 January 2007.
Update TRENDS in Ecology and Evolution Vol.22 No.3 115
www.sciencedirect.com