Download the Drosophila melanogaster species group

Biological Journal of the Linnean Society, 2002, 76, 21–37. With 2 figures
Phylogeny of a paradigm lineage: the Drosophila
melanogaster species group (Diptera: Drosophilidae)
VALERIE SCHAWAROCH1,2*
1
Department of Biology, City College, CUNY, New York, NY 10031, USA
Division of Invertebrate Zoology, American Museum of Natural History, Central Park West at 79th
Street, New York, NY10024-5192, USA
2
Received 28 May 2001; accepted for publication 3 January 2002
Although Drosophila melanogaster is a paradigm eukaryote for biology, relationships of this species and the other
174 species in the melanogaster species group are poorly explored and ambiguous. Gene regions of Cytochrome
oxidase II (mt:CoII), Alcohol dehydrogenase (Adh) and hunchback (hb) were sequenced and analysed phylogenetically to test prior hypotheses of relationships for the group based on chromosomes, morphology, and 28S rRNA gene
sequences. A simultaneous cladistic analysis of the three newly sequenced gene regions produced a single wellresolved phylogeny for 49 exemplar species representing eight subgroups. Monophyly of each of the ananassae,
melanogaster, montium, and takahashii subgroups is supported; the suzukii subgroup is polyphyletic. This phylogeny is consistent with variation in significant morphological structures, such as the male sex comb on the fore
tarsus. The broad range of morphological variation among these species is interpreted and the applicability to evolution and developmental investigations is discussed. This phylogeny facilitates comparative investigations, such
as gene family evolution, transposable element transmission, and evolution of morphological structures. © 2002
The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37.
ADDITIONAL KEYWORDS: Adh – DNA – hb – molecular phylogeny – mt:CoII – sex comb.
INTRODUCTION
THE D.
MELANOGASTER PARADIGM
Evolutionary studies rely on well-established phylogenies. Drosophila melanogaster traditionally and
currently serves as a model organism for virtually all
aspects of biology, but especially for genetics and
development (Lawrence, 1992; Kohler, 1994; Sullivan,
Ashburner & Hawley, 2000). This species, however,
is only one of 174 species within the melanogaster
species group. Techniques developed and information
gathered in D. melanogaster-based studies can be
expanded to the other closely related species. Previous
investigations employing Drosophila, such as the
evolution of gene families (Drosopoulou & Scouras,
1995; Inomata et al., 1997b; Inomata et al., 1997a),
*Current address: Division of Invertebrate Zoology, American
Museum of Natural History, Central Park West at 79th
Street, New York, NY 10024–5192, USA.
E-mail: [email protected]
chromosome structure (Mavragani-Tsipidou et al.,
1994; Drosopoulou & Scouras, 1995; Scouras, 1995),
and transposable element transmission (Tanda
et al., 1988; Daniels et al., 1990; Clark & Kidwell,
1997; Clark et al., 1998) would benefit from a wellestablished phylogeny. However, phylogenetic relationships for species within the melanogaster species
group are poorly understood.
PREVIOUS
HYPOTHESES OF RELATIONSHIPS
The melanogaster species group is one of eight
species groups within the subgenus Sophophora of the
genus Drosophila, and represents one of the largest
radiations of species within the genus Drosophila
(Bock, 1980). The obscura group has been established
as sister to the melanogaster group based on numerous morphological and biochemical investigations
(Sturtevant, 1942; Throckmorton, 1975; Powell &
DeSalle, 1995). Males in both groups possess a comb
of thick, sclerotized teeth (the ‘sex comb’) on the fore
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
21
22
V. SCHAWAROCH
Figure 1. Phylogenetic hypothesis generated from a most parsimonious simultaneous analysis of the three gene regions
(length = 1540 steps and CI = 0.35). The ingroup (i.e. the melanogaster group) exhibits monophyly where applicable for all
traditional taxonomic groupings, except the suzukii subgroup. Bold lines on the cladogram represent nodes where both
the Bremer support is ≥3 and bootstrap is >50%. Morphological structures (sex comb, epandrium and mid-tibia) discussed
in the text and that corroborate cladogram structure are illustrated. The evolutionary changes seen in the male sex comb
are labelled with hatched bars on the phylogeny. Sex comb orientation was denoted as either Hor. for horizontal, Obl. for
oblique or Long. for longitudinal. The number of foreleg tarsal segments containing sex comb teeth were listed as either
1 tar., 2 tar., or 3 tar.
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
THE MELANOGASTER SPECIES GROUP
tarsi, which is the most overt feature of the groups,
although scarcely developed in some species (Fig. 1).
This character has also been used to establish and
maintain the melanogaster group and its 12 named
subgroups.
There have been five major studies on the
melanogaster species group (Hsu, 1949; Okada, 1954;
Bock & Wheeler, 1972; Bock, 1980 and a regional
revision by Toda, 1991). Only the studies of Hsu (1949)
and Okada (1954) and Bock & Wheeler (1972) proposed relationships among the subgroups using
morphological characters, but without phylogenetic
bases. These classifications were presented when only
12–50% of the current species in the group were
known. Bock & Wheeler (1972) proposed that the
ananassae and montium species subgroups form one
lineage. Within the remaining subgroups they also
proposed a cluster of closely related subgroups (i.e.
eugracilis, ficusphila, suzukii and takahashii) which
could be distinguished by the small hooked bristles
on the mid-tibiae of the males and characters from
the male genitalia. Recent hypotheses provide some
support for this classification. Ashburner et al.
(1984), using chromosomes and morphology, discerned
three lineages: (1) an ananassae species subgroup;
(2) a montium species subgroup; and (3) a lineage
comprised of the elegans, eugracilis, ficusphila,
melanogaster, suzukii, and takahashii species subgroups. However, Ashburner et al. (1984) were unsure
how these lineages were interrelated or where to place
the remaining species subgroups.
The first major DNA sequencing studies of the
melanogaster group were by Pélandakis et al. (1991)
and Pélandakis & Solignac (1993) and contained
the greatest taxon sampling. Unfortunately, their
findings are controversial. Pélandakis et al. (1991) and
Pélandakis & Solignac (1993) used 28S rRNA
sequences with parsimony and Neighbour-Joining
analyses to reconstruct relationships within the genus
Drosophila and the subgenus Sophophora, including
21 species in the melanogaster group. The parsimony
analysis produced numerous trees whose consensus
lacked resolution; therefore, the proposed phylogeny
was based on Neighbour-Joining. The NeighbourJoining phylogeny had extensive paraphyly at various
levels (within the genus Drosophila, the subgenus
Sophophora, and the melanogaster group), contrary
to previous hypotheses based on morphological characters. With respect to the melanogaster group,
Pélandakis et al. (1991) and Pélandakis & Solignac
(1993) proposed three lineages: (1) the obscura and
fima groups allied with the ananassae subgroup; (2)
a montium subgroup lineage; and (3) a lineage comprised of the melanogaster subgroup plus the ‘Oriental’ (Asian) elegans, eugracilis, ficusphila, suzukii, and
takahashii subgroups. The Neighbour-Joining tree
23
had a low bootstrap value of 6 at the node uniting the
first with the second lineage.
Pélandakis et al. (1991) and Pélandakis & Solignac
(1993) attributed the lack of resolution with their
parsimony analyses, as well as the paraphyly and
poor bootstrap support in their Neighbour-Joining
phylogeny, to the small number of characters in the
28S gene supporting the ananassae + montium
subgroups node. The few 28S characters, they believe,
is a result of a rapid speciation event. Methodologically, Neighbour-Joining produces a single ‘resolved’
tree, however, its solutions are not considered optimal
and should not used for final phylogenetic hypotheses
(Hillis et al., 1996). A basal trichotomous relationship
for the subgroups is unlikely because distinctive morphological characters exist uniting some of the subgroups, such as hooked setae on the mid-tibia, and
presence of both a surstylar clasper and cercal clasper.
Rather, the lack of characters here probably reflects
the gene region used for the analyses. Within the
melanogaster group the D1 and D2 regions of 28S
rRNA are extremely invariant (see figs in Pélandakis
& Solignac, 1993; Schawaroch, 2000). In fact, the D3
expansion region of 28S rDNA evolves so slowly as to
resolve relationships among the holometabolous insect
orders – divergence events an order of magnitude
older than in drosophilids (see Whiting et al., 1997).
The final hypothesis of relationships proposed by
Pélandakis et al. (1991) and Pélandakis & Solignac
(1993) largely agrees with Ashburner et al. (1984)
in recognizing three distinct lineages within the
melanogaster group: (1) the ananassae subgroup; (2)
the montium subgroup; and (3) the melanogaster plus
Asian subgroups. There was no proposal for how these
lineages were interrelated or where to place the
remaining subgroups.
THE
PRESENT STUDY’S GOALS
The primary goal of the present study is to resolve
relationships within the melanogaster group by choosing more variable gene regions (i.e. Alcohol dehydrogenase [Adh], hunchback [hb], and Cytochrome oxidase
II [mt:CoII]) and by sampling more taxa (i.e. 43 species
representing eight subgroups [Fig. 1]). Morphological
characters employed by earlier studies were examined
in the context of this new molecular phylogeny. Relationships established by the molecular phylogeny will
be used to describe the evolution of sex comb transformation for the melanogaster species group.
MATERIAL AND METHODS
TAXON
SAMPLING
Flies were obtained from the National Drosophila
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
24
V. SCHAWAROCH
Species Resource Center at Bowling Green and from
D. Lachaise (Appendix A). A total of 49 taxa were used
in this study. The ingroup contains representative
taxa from eight of the 12 subgroups within the
melanogaster group. Outgroup taxa consist of six
species from the obscura group, two from each of the
subgroups of obscura, pseudoobscura and affinis in the
obscura group (Barrio et al., 1994).
GENE
REGIONS
One mitochondrial (mt:CoII) and two single copy
nuclear (Adh and hb) gene regions were chosen on
their ability to provide characters (synapomorphies) at
the taxonomic level of investigation (Schawaroch,
2000). All of these regions have conserved areas and
variable areas that are a source of characters at the
species level. Although each of these genes exhibits
population-level variability, species have their own
distinct sequences that are shared among populations
(Davis & Nixon, 1992). Any DNA sites found to be
heterozygous were coded as such. The DNA regions
chosen are used in a wide variety of taxa, especially
within Drosophila; thus facilitating comparisions with
previous studies (e.g. Beckenbach et al., 1993; Thomas
& Hunt, 1993; Baker & DeSalle, 1997). A simultaneous analysis of multiple unlinked gene regions was
employed to avoid problems of single gene phylogenies
(Doyle, 1992).
MOLECULAR
TECHNIQUES
Genomic DNA was prepared using single fly preps
(DeSalle et al., 1993). DNA was PCR amplified
using PE Taq polymerase with primers described
previously (Thomas & Hunt, 1993; Brower, 1994;
Baker & DeSalle, 1997; plus two new Adh primers
5¢-TGGGCGGCATTGGNYTNGAYAC-3¢
and
5¢AGCCAGGARTTGAAYTTRTG-3¢, Schawaroch, 2000).
PCR products were purified using Gene Clean II (Bio
101). The Adh fragment was cloned using Invitrogen’s
TA cloning kit for many of the taxa. Gene regions were
sequenced in both directions by either manual or
automated sequencing methods. Most of the Adh,
and mt:CoII sequences were generated manually. The
remainder including all the hb sequences were done
using an ABI 373 automated sequencer. Manual
sequencing of double stranded PCR products and
clones was done using 35S and United States Biochemical’s Sequenase version 2.0 DNA sequencing kit,
according to manufacturer’s instructions. Automated
sequencing of double stranded PCR product was
accomplished according to ABI Prism DNA sequencing
kit, purified by sephadex columns and run using
Applied Biosystems 373A machine and DNA sequence
protocols. Sequences were checked and corrected using
S E Q U E N C H E R 3.0 (Gene Codes Corp.) sequence
analysis software. Most of the DNA sequence was
generated by this study (GenBank accession numbers
– Adh: AF459744-AF459786; mt:CoII: AF461268AF461308; hb: AF461309-AF461356) with the following exceptions: affinis: mt:CoII (M95140); ambigua:
mt:CoII (M95145), Adh (X54813); bifasciata: mt:CoII
(M95147); melanogaster: mt:CoII (AF200828), Adh
(M11290), hb (Y00274); persimilis: mt:CoII (M95143),
Adh (M60997); pseudoobscura: mt:CoII (M95150),
Adh (M60989); teissieri: Adh (X54118); tolteca: mt:CoII
(M95147); and yakuba: mt:CoII (X00924), Adh
(X57376).
CHARACTER ASSIGNMENT
The Adh and mt:CoII DNA sequences of 290 and
384 bp, respectively, contained no insertions or
deletions; therefore, alignment was straightforward.
Homology assessment for hb was more complicated
because the total length of the hb sequence varied
from 513 bp in D. bifasciata to 456 bp in D. takahashii
and D. elegans. It was necessary to convert hb
nucleotide sequence to amino acid sequence for
recognition of homology (i.e. topological identity
sensu Brower & Schawaroch, 1996). Alignments were
performed on hb amino acid sequences using the
Clustal method in M E G A L I G N (D N A S TA R , version
1.02). A sensitivity analysis (Wheeler, 1995) was
performed varying the gap penalty from 8 to 30.
Because topological identity could not be established,
alignment-ambiguous sites were removed from the
analysis (Gatsey et al., 1994). The remaining gaps
were coded as a combination of question marks and
5th state depending upon the alignment context
(Appendix B).
MOLECULAR
CHARACTER ASSESSMENT
Character data were assessed for heterogeneity in
base composition. Total nucleotide composition for
each gene region was tested for heterogeneity using
HKY (Hasegawa et al., 1985) likelihood model under
P U Z Z L E (ver. 4.0.2, Strimmer & von Haeseler,
1999).
To quantify the congruence between each of the data
partitions and the combined analysis tree, the incongruence length difference (ILD) (Mickevich & Farris,
1981) was calculated for phylogenetically informative
characters. ILD values were tested for significance
(Farris et al., 1994; Farris et al., 1995) using the
partition-homogeneity test for 111 iterations with 10
random addition tree bisection-reconnection (TBR)
searches in PAUP* 4.0 (Beta ver. 4.0b2a-4.0b4a,
Swofford, 2002).
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
THE MELANOGASTER SPECIES GROUP
PHYLOGENETIC ANALYSES
A combined analysis (Kluge, 1989) of all three genes
for all the taxa was generated. The tree was rooted
with the six outgroup species chosen from the sister
taxon, the obscura group. Only informative characters
were used to generate trees and tree statistics. Heuristic tree searches were performed using PAUP* 4.0 (Beta
ver. 4.0b2a-4.0b4a, Swofford, 2002) with random addition of taxa, TBR branch swapping and repeated 20
times. The characters were given an equal weight of
one and run unordered.
CHARACTER
SUPPORT AT NODES
Support for nodes in the combined analyses was
evaluated by Bremer support (BS) (Bremer, 1988;
Bremer, 1994) and bootstrap (B) values (Felsenstein,
1985). BS values were calculated using A U T O D E C AY
(ver. 2.9.8, Eriksson, 1997). B analyses employed 1000
replicates with each replicate containing 10 heuristic
searches with random addition of taxa and TBR
branch swapping.
RESULTS
MOLECULAR
mt:CoII gene region has a strong A and T bias with
relatively little C and G nucleotide base content
(A = 33.8%, C = 13.6%, G = 15.2%, T = 37.4%). No correction was needed for this nucleotide bias because
each gene region passed a heterogeneity chi-square
test at the 5% level (P U Z Z L E ver. 4.0.2, Strimmer &
von Haeseler, 1999).
INCONGRUENCE LENGTH DIFFERENCE (ILD)
The data partitions are significantly heterogeneous
when compared to the combined data (P = 0.009). This
paper, however, employs a combined data analysis
for phylogeny reconstruction based on the following
reasons. Sources of evidence (i.e. characters) should be
varied, thereby negating the problem of single character or gene phylogenies (or homogenized data) (e.g.
Doyle, 1992). By not including all the data, resolution
could be lost especially if data partitions contribute
information for different levels of the analysis (e.g.
Hillis, 1987). Studies have demonstrated that
simultaneous analyses provide greater resolution
(Olmstead & Sweere, 1994; Miller et al., 1997;
Remsen & DeSalle, 1998).
CHARACTER ASSESSMENT
Approximately 30% of all the nucleotides sequenced
were phylogenetically informative (Table 1). All the
DNA regions used are protein coding. As expected,
third position sites contributed the greatest number of
phylogenetically informative characters, and second
position sites contributed the least.
BASE
COMPOSITION
Within separate gene regions the total nucleotide base
composition exhibited biases. Nuclear regions have a
bias in favour of the C nucleotide base (Adh: A = 24.6%,
C = 30.2%, G = 25.8%, T = 19.5%; hb: A = 25.5%,
C = 36.7%, G = 23.9%, T = 14%). As in other insect
mitochondrial studies (Clary & Wolstenholme, 1985;
DeSalle et al., 1987; Liu & Beckenbach, 1992), the
25
TOTAL EVIDENCE ANALYSIS
The simultaneous analysis (mt:CoII + Adh + hb)
resulted in a well resolved, single most parsimonious
(SMP) cladogram (Length = 1540 steps, CI = 0.35)
(Fig. 1). The CI value is within the standardized estimate for the number of taxa (Sanderson & Donoghue,
1989). In this tree monophyly was strongly supported
for the melanogaster group (BS = 36; B = 100%).
Monophyly was also supported for the ananassae,
melanogaster, montium and takahashii subgroups
with BS = 9, 8, 14, 1 and B = 98%, 90%, 99%, <50%,
respectively. As the small elegans and ficusphila subgroups were each represented by a single taxon in this
study, monophyly could not be tested. Monophyly is
untestable for the monotypic eugracilis subgroup. In
this analysis, the suzukii subgroup was polyphyletic.
For the suzukii subgroup representatives: D. mimet-
Table 1. Contributions of the various gene regions and nucleotide positions to the number of phylogenetically informative
characters in the dataset
Phylogenetically informative
Data
Total characters
Phylogenetically
informative characters
All three gene regions
Adh
hb
mt:Coll
1108
290
434
384
342
92
138
112
First position
Second position
Third position
52
18
16
18
21
8
12
1
267
66
108
93
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
26
V. SCHAWAROCH
ica was basal in the takahashii clade (BS = 3; B = 59%),
D. biarmipes was basal in the clade containing the
melanogaster and eugracilis subgroups (BS = 2;
B < 50%), and D. lucipennis forms a clade with
D. elegans (BS = 12; B = 99%). This analysis also
detected three major lineages within the melanogaster
group: ananassae subgroup, the montium subgroup
and the melanogaster + Asian subgroups, confirming
the classification of Ashburner et al. (1984). The
ananassae and montium subgroups appear to be sister
taxa, although there is weak BS and B support
(BS = 1; B < 50%).
DISCUSSION
Phylogenetic hypotheses are constrained by the taxa
and characters sampled. The importance of taxon
choice and its influence on cladogram structure has
been demonstrated (Lecointre et al., 1993; Graybeal,
1998; Hillis, 1998; Poe, 1998). To obtain an adequate
DNA sample for small species such as drosophilids
the whole specimen must be sacrificed. Fortunately,
a number of drosophilid species are maintained in
laboratory cultures. These cultures, however, limit the
scope of the molecular studies. Most, but not all, of the
available stocks have been included in the present
study, which included 25% of the currently known
species in the melanogaster group – the most comprehensive taxon sampling for the melanogaster group
for any biochemical investigation thus far. In some
instances the resulting phylogenetic hypotheses may
seem to contradict traditional views or appear poorly
supported, however, they are actually in agreement
and are corroborated by other types of data (e.g.
morphological and electrophoretic) and methods of
analysis. This corroboration indicates stability of the
hypothesis of relationships proposed here.
THE
SPECIES SUBGROUPS
The ananassae subgroup
The ananassae subgroup has been characterized by
the presence of a cercal clasper and a surstylar clasper
with two sets of teeth (Figs 1 and 2). For members of
this subgroup, D. varians is morphologically unique in
lacking the cercal clasper and possessing a cercal plate
with bristles similar to species of the suzukii subgroup
(Fig. 1). Drosophila varians has been included within
the ananassae subgroup based on chromosomal characters (Bock & Wheeler, 1972). In the present study
six species were chosen to represent the ananassae
subgroup, three from the ananassae complex
(D. ananassae, pallidosa and phaeopleura), one from
the bipectinata complex (D. malerkotliana), one from
the ercepeae complex (D. ercepeae), and one unassigned species, D. varians. The SMP cladogram supported monophyly for the ananassae subgroup with
very high BS and B values (BS = 9, and B = 98%).
Within the ananassae subgroup clade, species of the
ananassae complex form a cluster. As the remaining
species complexes have only single representatives it
can not be determined if they are natural groups. It is
interesting to note that despite previous studies questioning its inclusion, the morphologically aberrant
D. varians does turn out to be nested within the
ananassae subgroup.
Figure 2. Foreleg and male periphallic structures; terms refer to structures discussed in the text.
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
THE MELANOGASTER SPECIES GROUP
The melanogaster subgroup
The melanogaster subgroup has been well investigated
and much data has been used to indicate its monophyly (Lachaise et al., 1988). This study sampled
the familiar D. melanogaster and the two species of
the yakuba complex, D. yakuba and D. teissieri. The
melanogaster subgroup forms a well-supported clade
(BS = 8 and B = 90%) located in a relatively more
derived position within various Asian species and
subgroups. The yakuba complex has even greater
Bremer and bootstrap values (BS = 16 and B = 100%).
The SMP cladogram agrees with the current hypothesis of relationships for the yakuba complex + D.
melanogaster.
The montium subgroup
The montium subgroup is monophyletic with very high
BS and B values (BS = 14, and B = 99%) at the basal
node. A monophyletic montium subgroup is in agreement with morphological and biochemical studies for
which various synapomorphies have been proposed
(e.g. Bock & Wheeler, 1972; Ashburner et al., 1984;
Scouras, 1995). This contrasts with Tsacas & David
(1978) who felt that the montium subgroup could not
be monophyletic due to its enormous size and widespread distribution, criteria not applicable to defining
monophyletic groups. Toda (1991) removed three
species (i.e. D. rhopaloa, palmata and longissima)
from the montium subgroup, for which he established
the rhopaloa and longissima subgroups. Unfortunately, none of these species or other representatives
of these subgroups were included in the present study.
One way to assess this classification in lieu of sequencing is to assess diagnostic value of morphological characters, as determined by total evidence of the analysed
taxa. The two species in the longissima subgroup
possess a sex comb identical to ones in the montium
subgroup, which consists of two long, longitudinally
arranged rows of numerous teeth on tarsomeres 1 and
2. As this feature corroborates the molecular phylogeny presented here, the longissima subgroup may
either be a complex within montium subgroup or a
sister taxon to the montium subgroup. Many morphological features of the rhopaloa subgroup are variable
and its monophyly is presently questionable. A
thorough discussion of the complexes and relationships within this large complicated subgroup is provided in Schawaroch (2000). It can be noted that: (1)
D. barbarae is not a member of the clade that contains
representatives of either jambulina or kikkawai complexes; and (2) much of the resolution within the
montium subgroup is supported by BS-values less
than or equal to 2 and B values less than 50%.
The takahashii subgroup
The SMP cladogram supports monophyly for the taka-
27
hashii subgroup, even though BS and B values are low
(BS = 1 and B < 50%). The takahashii subgroup species
grouped as two complexes: paralutea + prostipennis
and takahashii + lutescens. This differs from the
hypothesized affinity of D. lutescens, paralutea,
pseudotakahashii, takahashii and trilutea based on
hybridization tests (Bock & Wheeler, 1972; Watanabe
& Kawanishi, 1983; Lemeunier et al., 1986). A morphological character (i.e. number of rows of male sex
comb teeth on the second tarsal segment) (Figs 1 and
2) and electrophoresis of allozymes by Parkash et al.
(1994) corresponds with the division of the takahashii
subgroup seen in the current study’s molecular cladogram. Although included within the melanogaster +
Asian subgroups clade, the takahashii subgroup is not
the sister taxon to the melanogaster subgroup as Bock
(1980) hypothesized.
The suzukii subgroup
Members of this subgroup apparently exhibit the greatest disparity of morphological characters, particularly
for sex comb, phallic and periphallic structures, which
accounts for polyphyly of the ‘subgroup.’ The putative
synapomorphies for this ‘subgroup’ are generalized
male genitalic characters, such as surstylar clasper
with several sets of distinctly different teeth, cercal
plate with lower bristles differentiated from upper bristles, and large posterior paramere (Toda, 1991). For
these reason, monophyly of the suzukii ‘subgroup’ has
been questioned (Bock & Wheeler, 1972; Bock, 1980;
Toda, 1991). In the SMP cladogram the suzukii subgroup was polyphyletic. The three suzukii ‘subgroup’
representatives of D. mimetica, biarmipes and lucipennis used in this study exhibit the complete range in
variation with respect to sex comb structure.
The SMP molecular-based analysis places each of
these three representative species in a separate clade
that corresponds well with their sex comb morphology.
Drosophila mimetica is sister to the takahashii clade
(BS = 3; B = 59%), and all taxa have similar sex
combs – a horizontal row each on the first and second
tarsal segments. Drosophila biarmipes is the most
basal member in a clade containing the melanogaster
and eugracilis subgroups (BS = 2; B < 50%). All taxa
have a sex comb located on the first tarsal segment.
However, the D. biarmipes and the melanogaster
subgroup sex comb has an oblique orientation,
whereas, the eugracilis sex comb contains two teeth of
either oblique or longitudinal orientation. Drosophila
lucipennis forms a clade with D. elegans (BS = 12;
B = 99%). Drosophila lucipennis has completely lost
the sex comb. In D. elegans, the sex comb is a series
of horizontal rows along the first three tarsal segments. Bock & Wheeler (1972: 27–28) established a
subgroup for D. elegans alone, because it “. . . differs
substantially in the structure of the male genitalia
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
28
V. SCHAWAROCH
[from suzukii subgroup].” Only the D. lucipennis plus
D. elegans clade has high BS and B values. Low BS
and B values were also seen for the clades containing
D. mimetica and D. biarmipes (BS = 2; B < 50%), and
the sister to this clade, which contains D. lucipennis
(BS = 3; B = 63%). Further work on the status and/or
redefinition of the suzukii ‘subgroup’ is indicated.
RELATIONSHIPS AMONG
CLADES
The three lineages
In the SMP cladogram the melanogaster group is subdivided into three major clades: the ananassae subgroup, the montium subgroup, and the melanogaster
+ Asian subgroups. This agrees with Ashburner et al.
(1984). Ashburner et al. (1984), however, did not
hypothesize any relationships among these lineages.
My hypothesis also agrees with Pélandakis et al.
(1991), who interpreted their findings to support the
three lineages of Ashburner et al. (1984) with the
exception of the obscura and fima groups within
the melanogaster group. The SMP cladogram hypothesized relationships among the three lineages as:
(ananassae, montium) melanogaster + Asian
subgroups.
The ananassae and montium subgroups are
sister taxa
The SMP cladogram proposes the ananassae and
montium subgroups as sister taxa. The node supporting this relationship has low BS and B values (BS = 1
and B < 50%). Despite the weak support, the sister
relationship between the ananassae and montium
subgroups also has a morphological basis, indicated
even by the early studies of Hsu (1949) and Okada
(1954). Bock & Wheeler (1972) hypothesized that the
ananassae and montium subgroups formed a lineage
based on the presence of both a surstylar clasper and
cercal clasper (Figs 1 and 2). Ashburner et al. (1984)
hypothesized no resolution among the three lineages.
Pélandakis et al. (1991), and even Pélandakis &
Solignac (1993), presented the relationship: subgenus
Drosophila [melanogaster + Asian (montium, ananassae + fima + obscura)]. Despite the paraphyly and the
lack of characters within Pélandakis et al. (1991) and
Pélandakis & Solignac (1993) studies, the ananassae
and montium subgroups are still seen as having a
greater affinity.
phila. The support for these relationships varies but
they have relatively low BS (< 3) and B (< 50%) values.
Okada (1964) and Bock & Wheeler (1972) each felt
that the eugracilis, ficusphila, suzukii and takahashii
subgroups had a close affinity due to the hooked setae
on the mid-tibiae of males and other characters of the
male genitalia. The SMP cladogram does not exactly
support this hypothesis because the melanogaster
and elegans subgroups are nested within that clade.
Mapping the hooked setae character on the SMP
cladogram (i.e. eugracilis, ficusphila, suzukii, and
takahashii representatives are coded for presence,
all other taxa absence) increases the length of the
tree by three steps and does not affect the CI value
(343 informative characters, vs. the previous 342,
therefore L = 1543, CI = 0.35). According to the SMP
cladogram, the presence of hooked setae on the
mid-tibia may have evolved once at the base of the
melanogaster + Asian subgroups clade and was lost
twice: once at the node for the melanogaster subgroup
clade and a second time at the terminal for the elegans
subgroup. Therefore, this character may actually be a
synapomorphy for the melanogaster + Asian subgroups
lineage.
The SMP cladogram disagrees with previous
hypotheses. Hsu (1949) and Okada (1954) placed the
suzukii subgroup at the base of the melanogaster
group, whereas within the SMP cladogram none of the
suzukii representatives are basal either within the
melanogaster group or the melanogaster + Asian
subgroups clade. Contrary to Bock (1980), the SMP
cladogram places melanogaster as sister to the eugracilis subgroup rather than the takahashii subgroup.
The electrophoretic and hybridization (breeding)
studies of Kim & Lee (1991), Kim et al. (1992), Lee et
al. (1993), and Lee et al. (1994) hypothesized a hierarchy of (melanogaster, takahashii) suzukii which was
not supported by the SMP cladogram. Also in contrast
to the SMP cladogram was Nigro et al. (1991) mitochondrial DNA-based scheme of relationships for the
melanogaster + Asian subgroups as: D. eugracilis
(D. takahashii (melanogaster subgroup)). However,
the SMP cladogram’s subdivisions within the
melanogaster + Asian subgroups clade best explains
changes in sex comb morphology (see previous discussion for suzukii subgroup).
CONCLUSION
Relationships within the melanogaster and
Asian subgroups
The melanogaster + Asian subgroups clade seems well
supported with a BS value of 3 and a B value of 85%.
The relationships of the subgroups within this clade
can be summarized as: ((((melanogaster, eugracilis)
suzukii) (takahashii, suzukii)) (elegans, suzukii)) ficus-
POTENTIAL
FOR COMPARATIVE DEVELOPMENTAL
EVOLUTION OF MORPHOLOGICAL STRUCTURES
Phylogenies can be used to establish or to test evolutionary hypotheses explaining variation exhibited
by a group of organisms. Within the melanogaster
species group the male sex comb is a well-documented,
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
THE MELANOGASTER SPECIES GROUP
highly variable structure. A male sex comb, possessed in varying degrees, is a synapomorphy for the
melanogaster and obscura groups within the subgenus
Sophophora and unique within the family Drosophilidae (Grimaldi, 1990). Within the melanogaster species
group the sex comb character has been used to define
species subgroups (e.g. Bock & Wheeler, 1972; Bock,
1980; Toda, 1991). The male sex comb functions in
species mate recognition, courtship and copulation
(Spieth & Ringo, 1983). By mapping the sex comb
character on the current molecular phylogeny a
hypothesis can be generated describing sex comb
structural transformation.
The sex comb varies in general length, number of
tarsal segments and orientation (Fig. 1). Mapping the
sex comb types on the SMP phylogeny illustrates the
usefulness of this character for diagnosing the subgroups, as it has been used in previous studies (Fig. 1).
Loss of sex comb, seen here in D. lucipennis, also
occurs in five other melanogaster group species belonging to four subgroups. Homology statements should
ideally not be based on loss (absence) of a character;
therefore, other character information, including
molecular data, will be necessary for phylogenetic
reconstruction.
Evolution of the sex comb among groups and
subgroups is best understood on the basis of two
characters – orientation and number of tarsal segments. When the orientation of the sex comb was
mapped (MacClade ver. 3.01, Maddison & Maddison,
1992) on to the SMP phylogeny using the accelerated
transformation (ACCTRAN) model of character evolution there were minimally two origins for longitudinal
orientation (one for ficusphila and the other for the
montium subgroup), or minimally two origins for a
horizontal orientation (one at the node shared by the
takahashii subgroup-mimetica clade + the elegans subgroup and the other for the ananassae subgroup).
There was a reversal in sex comb orientation to the
plesiomorphic condition of oblique for the biarmipeseugracilis-melanogaster subgroup clade. A sex comb
occupying two tarsal segments is the plesiomorphic
condition for the melanogaster group on the SMP phylogeny, which is not surprising because the outgroup
(the obscura species group) shows this state. Employing an ACCTRAN evolutionary hypothesis produces two instances where the sex comb was reduced
to a single tarsal segment (biarmipes-eugracilismelanogaster clade and the other in the derived
montium subgroup species, D. nikananu. Drosophia
nikananu is a member of a species complex characterized by short sex combs.) Due to the similarity of
the sex combs to those of the melanogaster subgroup,
the nikananu complex has been proposed as most
basal within the montium subgroup, linking the
montium with the melanogaster subgroup (Tsacas,
29
1979; Tsacas, 1984; Tsacas & Chassagnard, 1992).
However, it is important to note that this ‘similar’
character differs in orientation and is therefore structurally and phylogenetically convergent. A sex comb
occupying three tarsal segments occurs in two lineages: the elegans subgroup and the ananassae subgroup (with a reversal appearing in D. malerkotliana
whose sex comb covers only two tarsal segments).
Future studies might explore comparative developmental mechanisms involved in sex comb formation.
Studies exist which explain the relation between
genes and phenotypic expression in Drosophila (Liu
et al., 1996; True et al., 1997; Rutherford & Lindquist,
1998; Zeng et al., 2000). Many genes are involved in
sex comb development, including Sex combs reduced
(Scr) and cramped (crm). Scr affects the number of
teeth in a sex comb (Pattatucci et al., 1991) – a feature
that varies at the species level within subgroups of the
melanogaster group. The crm gene causes proximaldistal transformations that produce sex comb teeth to
be present on second and third tarsal segments
(Yamamoto et al., 1997) – a feature used in some cases
to define species subgroups. Reliable evolutionary
hypotheses of important structures (e.g. sex comb and
Balbiani rings) or biochemical entities (e.g. gene
families and transposable elements) are possible only
with informative, stable phylogenies like the one
presented here for the melanogaster group.
ACKNOWLEDGEMENTS
NSF Doctoral Dissertation Improvement Grant (NSF
DEB 9423508) and two City College Biology Dissertation Grants funded this research. AMNH-CUNY Doctoral Training Program Fellowship provided a stipend.
Research was conducted at the AMNH Molecular
Systematics Laboratory. I am most grateful to the
following individuals: Rob DeSalle who supported the
molecular work, Dave Grimaldi for tutelage regarding
Drosophila, to Carole Griffiths and Gail Simmons
for providing informative commentary, and Steve
Thurston for the figures. I would like to thank Michael
Ashburner and two anonymous reviewers who suggested changes that improved this paper.
REFERENCES
Ashburner M, Bodmer M, Lemeunier F. 1984. On the
evolutionary relationships of Drosophila melanogaster.
Developmental Genetics 4: 295–312.
Baker RH, DeSalle R. 1997. Multiple sources of molecular
characters and the phylogeny of Hawaiian drosophilids.
Systematic Biology 46: 654–673.
Barrio E, Latorre A, Moya A. 1994. Phylogeny of the
Drosophila obscura species group deduced from mitochondr-
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
30
V. SCHAWAROCH
ial DNA sequences. Journal of Molecular Evolution 39: 478–
488.
Beckenbach AT, Wei YW, Liu H. 1993. Relationships in the
Drosophila obscura species group, inferred from mitochondrial cytochrome oxidase II sequences. Molecular Biology
and Evolution 10: 619–634.
Bock IR. 1980. Current status of the Drosophila melanogaster
species-group (Diptera). Systematic Entomology 5: 341–356.
Bock IR, Wheeler MR. 1972. The Drosophila melanogaster
species group. University of Texas Publications: 7213: 1–102.
Bremer K. 1988. The limits of amino acid sequence data in
angiosperm phylogenetic reconstruction. Evolution 42: 795–
803.
Bremer K. 1994. Branch support and tree stability. Cladistics
10: 295–304.
Brower AVZ. 1994. Phylogeny of Heliconius butterflies
inferred from mitochondrial DNA sequences (Lepidoptera:
Nymphalidae). Molecular Phylogenetics and Evolution 3:
159–174.
Brower AVZ, Schawaroch V. 1996. Three steps of homology
assessment. Cladistics 12: 265–272.
Clark JB, Kidwell MG. 1997. A phylogenetic perspective on
P transposable element evolution in Drosophila. Proceedings
of the National Academy of Sciences, U.S.A. 94: 11428–
11433.
Clark J, Kim PC, Kidwell MG. 1998. Molecular evolution of
P transposable elements in the genus Drosophila. 3. The
melanogaster species group. Molecular Biology and Evolution 15: 746–755.
Clary DO, Wolstenholme DR. 1985. The mitochondrial DNA
molecule of Drosophila yakuba. Nucleotide sequence, gene
organization, and genetic code. Journal of Molecular Evolution 22: 252–271.
Danforth BN, Sauquert H, Packer L. 1999. Phylogeny of
the bee genus Halictus (Hymenoptera. Halictidae) based on
parsimony and likelihood analyses of nuclear EF-1 alpha
sequence data. Molecular Phylogenetics and Evolution 13:
605–618.
Daniels SB, Chovnick A, Boussy I. 1990. Distribution of
hobo transposable elements in the genus Drosophila.
Molecular Biology and Evolution 7: 589–606.
Davis JI, Nixon KC. 1992. Populations, genetic variation,
and the delimitation of phylogenetic species. Systematic
Biology 41: 421–435.
Dayhoff MO. 1978. Atlas of protein sequence and structure.
Silver Spring, Maryland: National Biomedical Research
Foundation.
DeSalle R, Freedman T, Prager EM, Wilson AC. 1987.
Tempo and mode of sequence evolution in mitochondrial
DNA of Hawaiian Drosophila. Journal of Molecular Evolution 26: 157–164.
DeSalle R, Williams AK, George M. 1993. Isolation and
characterization of animal mitochondrial DNA. In: Zimmer
EA, White TJ, Cann RL, Wilson AC, eds. Methods in enzymology. London: Academic Press, 176–204.
Doyle JJ. 1992. Gene trees and species trees: molecular
systematics as one-character taxonomy. Systematic Botany
17: 144–163.
Drosopoulou E, Scouras ZG. 1995. The B-tubulin gene
family evolution in the Drosophila montium subgroup of the
melanogaster species group. Journal of Molecular Evolution
41: 293–298.
Eriksson T. 1997. Autodecay 2.9.8. Stockholm: Stockholm
University, Botaniska Institutionen.
Farris JS, Kallersjo M, Kluge AG, Bult C. 1994. Testing
significance of congruence. Cladistics 10: 315–320.
Farris JS, Kallersjo M, Kluge AG, Bult C. 1995. Constructing a significance test for incongruence. Systematic
Biology 44: 570–572.
Felsenstein J. 1985. Confidence limits of phylogenies: an
approach using the bootstrap. Evolution 39: 783–791.
Gatsey J, DeSalle R, Wheeler WC. 1994. Alignmentambiguous nucleotide sites and the exclusion of systematic
data. Molecular Phylogenetics and Evolution 2: 152–157.
Graybeal A. 1998. Is it better to add taxa or characters to a
difficult phylogenetic problem? Systematic Biology 47: 9–17.
Grimaldi DA. 1990. A. phylogenetic, revised classification
of genera in the Drosophilidae (Diptera). Bulletin of the
American Museum of Natural History 197: 1–139.
Hasegawa M, Kishino H, Yano T. 1985. Dating of the
human-ape splitting by a molecular clock of mitochondrial
DNA. Journal of Molecular Evolution 21: 160–174.
Hillis DM. 1987. Molecular versus morphological approaches
to systematics. Annual Review of Ecology and Systematics
18: 23–42.
Hillis DM. 1998. Taxonomic sampling, phylogenetic accuracy
and investigator bias. Systematic Biology 47: 3–8.
Hillis DM, Mable BK, Moritz C. 1996. Applications of molecular systematics. The state of the field and a look to the
future. In: Hillis DM, Moritz C, Mable BK, eds. Molecular
systematics 2nd edn. Sunderland, MA: Sinauer Associates
Inc., 515–543.
Hsu TC. 1949. The external genital apparatus of male
Drosophilidae in relation to systematics. University of Texas
Publications 4920: 80–142.
Inomata N, Tachida H, Yamazaki T. 1997a. Molecular evolution of the Amy multigenes in the subgenus Sophophora of
Drosophila. Molecular Biology and Evolution 14: 942–950.
Inomata N, Tachida H, Yamazaki T. 1997b. Molecular evolution of the Amy multigenes in the subgenus Sophophora of
Drosophila. Molecular Biology and Evolution 14: 1338.
Kim NW, Lee TJ. 1991. Genetic relationships among the eight
species of the Drosophila melanogaster species group by
allozyme analysis. Korean Journal of Genetics 13: 297–309.
Kim NW, Lee TJ, Song ES. 1992. Evolutionary genetic study
on the eight species of the Drosophila melanogaster group
from Korea: reproductive isolation and protein analysis.
Korean Journal of Zoology 35: 211–218.
Kluge AG. 1989. A concern for evidence and a phylogenetic
hypothesis of relationships among Epicrates (Boidae, Serpentes). Systematic Zoology 38: 7–25.
Kohler RE. 1994. Lords of the fly: Drosophila genetics and the
experimental life. Chicago: University of Chicago Press.
Lachaise D, Cariou M-L, David JR, Lemeunier F, Tsacas
L, Ashburner M. 1988. Historical biogeography of the
Drosophila melanogaster species subgroup. In: Hecht MK,
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
THE MELANOGASTER SPECIES GROUP
Wallace B, Prance GT, eds. Evolutionary biology. New York:
Plenum Press, 159–225.
Lawrence PA. 1992. The making of a fly: the genetics of
animal design. Blackwell Scientific Publications, Oxford.
Lecointre GH, Philippe H, Vân Lê HL, Guyader HL. 1993.
Species sampling has a major impact on phylogenetic inference. Molecular Phylogenetics and Evolution 2: 205–224.
Lee TJ, Hong KJ, Kim NW. 1994. Genetic relationships
and protein variations during development within the
Drosophila melanogaster species group. II. Analysis of
soluble protein by 2DE. Korean Journal of Zoology 37:
249–254.
Lee TJ, Hong KJ, Song ES. 1993. Genetic relationships
and protein variations during development within the
Drosophila melanogaster species group. I. Analysis of soluble
protein by SDS-PAGE. Korean Journal of Genetics 15:
269–276.
Lemeunier F, David JR, Tsacas L, Ashburner M. 1986.
The melanogaster species group. In: Ashburner M,
Carson HL, Thompson JN, eds. The genetics and biology of
Drosophila vol. 3e. London: Academic Press, 147–256.
Liu H, Beckenbach AT. 1992. Evolution of the mitochondrial cytochrome oxidase II gene among ten orders of insects.
Molecular Phylogenetics and Evolution 1: 41–52.
Liu J, Mercer JM, Stam LF, Gibson GC, Zeng Z-B, Laurie
CC. 1996. Genetic analysis of a morphological shape difference in the male genitalia of Drosophila simulans and D.
mauritiana. Genetics 142: 1129–1145.
Maddison DR, Maddison WP. 1992. MacClade: 3.01. Analysis of phylogeny and Character Evolution. Sunderland,
Massacheusetts: Sinauer Associates.
Mavragani-Tsipidou P, Zambetaki A, Kleanthous K,
Pangou E, Scouras ZG. 1994. Cytotaxonomic differentiation of the Afrotropical Drosophila montium subgroup:
D. diplacantha and D. seguyi. The major role of reverse
tandem duplications. Genome 37: 935–944.
Mickevich M, Farris S. 1981. The implications of congruence
in Menidia. Systematic Biology 30: 351–370.
Miller JS, Brower AVZ, DeSalle R. 1997. Phylogeny of the
neotropical moth tribe Josiini (Notodontidae: Dioptinae).
Evidence from DNA sequences and morphology. Biological
Journal of the Linnean Society 60: 297–316.
Nigro L Solignac M, Sharp PM. 1991. Mitochondrial
sequence divergence in the melanogaster and Oriental
species subgroups of Drosophila. Journal of Molecular
Evolution 33: 156–162.
Okada T. 1954. Comparative morphology of the drosophilid
flies. I. Phallic organs of the melanogaster group. Kontyû
22: 36–46.
Okada T. 1964. Drosophilidae (Diptera) of Southeast Asia collected by the Thai-Japanese Biological Expedition 1961–62.
In: Kira T, Umesao T, eds. Nature and life in southeast Asia.
Kyoto: Fauna and Flora Research Society, 439–466.
Olmstead RG, Sweere JA. 1994. Combining data in phylogenetic systematics: An empirical approach using three
molecular data sets in the Solanaceae. Systematic Biology
43: 467–481.
Parkash R, Iyoutsna, Vanda [sic]. 1994. Allozyme phy-
31
logeny of five species of takahashii species subgroup of
Drosophila. Korean Journal of Genetics 16: 187–196.
Pattatucci AM, Ottenson DC, Kaufman T. 1991. A functional and structural analysis of the sex combs reduced locus
of Drosophila melanogaster. Genetics 129: 423–441.
Pélandakis M, Higgins DG, Solignac M. 1991. Molecular
phylogeny of the subgenus Sophophora of Drosophila
derived from the large subunit of ribosomal RNA sequences.
Genetica 84: 87–94.
Pélandakis M, Solignac M. 1993. Molecular phylogeny of
Drosophila based on ribosomal RNA sequences. Journal of
Molecular Evolution 37: 525–543.
de Pinna MCC. 1991. Concepts and tests of homology in the
cladistic paradigm. Cladistics 7: 367–394.
Platnick NI, Griswold CE, Coddington JA. 1991. On
missing entries in cladistic analysis. Cladistics 7: 337–
343.
Poe S. 1998. Sensitivity of phylogenetic estimation to taxonomic sampling. Systematic Biology 47: 18–31.
Powell JR, DeSalle R. 1995. Drosophila molecular phylogenies and their uses. Evolutionary Biology 28: 87–138.
Remsen J, DeSalle R. 1998. Character congruence of multiple data partitions and the origin of the Hawaiian Drosophilidae. Molecular Phylogenetics and Evolution 9: 225–235.
Rutherford SL, Lindquist S. 1998. Hsp90 as a capacitor for
morphological evolution. Nature 396: 336–342.
Sanderson MJ, Donoghue MJ. 1989. Patterns of variation
in levels of homoplasy. Evolution 43: 1781–1795.
Schawaroch VA. 2000. Molecular phylogeny of the Drosophila
melanogaster species group with special emphasis on the
montium subgroup. PhD Thesis, The City University of New
York. XII, 339.
Scouras ZG. 1995. The Drosophila montium subgroup
species: recent cytogenetic, molecular, development and evolutionary studies. Bios (Thessaloniki) 3: 125–158.
Spieth HT, Ringo JM. 1983. Mating behavior and sexual
isolation in Drosophila vol 3c. In: Ashburner M, Carson HL,
Thompson JN, eds. The genetics and biology of Drosophila.
New York: Academic Press, 223–341.
Strimmer K, von Haeseler A. 1999. PUZZLE. 4.0.2.
Sturtevant AH. 1942. The classification of the genus
Drosophila, with descriptions of nine new species.
University of Texas Publications 4213: 5–51.
Sullivan W, Ashburner M, Hawley RS. 2000. Drosophila
protocols. Cold Spring Harbor, New York: Cold Spring
Harbor Laboratory Press.
Swofford DL. 2000. PAUP*. Phylogenetic Analysis Using
Parsimony (*and other methods), version 4. Sunderland, MA:
Sinauer Associates, Inc.
Tanda S, Shrimpton AE, Ling-Ling C, Itayama H,
Matsubayashi H, Saigo K, Tobari YN, Langley CH.
1988. Retrovirus-like features and site specific insertions of
a transposable element, Tom, in Drosophila ananassae.
Molecular Genetics 214: 405–411.
Thomas RH, Hunt JA. 1993. Phylogenetic relationships in
Drosophila. A conflict between molecular and morphological
data. Molecular Biology and Evolution 10: 362–374.
Throckmorton LH. 1975. The phylogeny, ecology, and geog-
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
32
V. SCHAWAROCH
raphy of Drosophila. In: King RC, ed. Handbook of genetics.
New York: Plenum, 421–469.
Toda MJ. 1991. Drosophilidae (Diptera) in Myanmar (Burma)
VII. The Drosophila melanogaster species-group, excepting
the D. montium species-subgroup. Oriental Insects 25: 69–
94.
True JR, Liu J, Stam LF, Zeng Z-B, Laurie CC. 1997.
Quantitative genetic analysis of divergence in male secondary sexual traits between Drosophila simulans and
Drosophila mauritiana. Evolution 51: 816–832.
Tsacas L. 1979. Contribution des données africaines à la
compréhension de la biogéographie et de l’évolution du
sous-genre Drosophila (Sophophora) Sturtevant (Diptera,
Drosophilidae). Compté rendu des séances de la Société de
biogéographie 48: 29–51.
Tsacas L. 1984. Nouvelles données sur la biogéographie et
l’évolution du groupe Drosophila melanogaster en Afrique.
Description de six nouvelles espèces. (Diptera, Drosophilidae). Annales de la Societe Entomologique de France (N.S.)
20: 419–438.
Tsacas L, Chassagnard M-T. 1992. Le complex Drosophila
nikananu: description d’une nouvelle espèce africaine et
analyse de quelques caracters morphologiques du groupe
melanogaster (Diptera, Drosophilidae). Nouvelle Revue
d’Entomologie (N. S.) 8: 385–398.
Tsacas L, David J. 1978. Systematics and biogeography of
the Drosophila kikkawai-complex, with descriptions of new
species (Diptera, Drosophilidae). Annales de la Societe
Entomologique de France (N.S.) 13: 675–693 (published in
1977).
Watanabe TK, Kawanishi M. 1983. Stasipatric speciation
in Drosophila. Japanese Journal of Genetics 58: 269–274.
Wheeler WC. 1993. The triangle inequality and character
analysis. Molecular Biology and Evolution 10: 707–712.
Wheeler WC. 1995. Sequence alignment, parameter sensitivity, and the phylogenetic analysis of molecular data. Systematic Biology 44: 321–331.
Whiting MF, Carpenter JC, Wheeler QD, Wheeler WC.
1997. The Strepsiptera problem: phylogeny of the
holometabolous insect orders inferred from 18S and 28S
ribosomal DNA sequences and morphology. Systematic
Biology 46: 1–68.
Yamamoto Y, Girard F, Bello B, Affolter M, Gehring WJ.
1997. The cramped gene of Drosophila is a member of the
Polycomb-group, and interacts with mus209, the gene encoding proliferating cell nuclear antigen. Development 124:
3385–3394.
Zeng Z-B, Liu J, Stam LF, Kao C-H, Mercer JM, Laurie
CC. 2000. Genetic architecture of a morphological shape
difference between two Drosophila species. Genetics 154:
299–310.
APPENDIX A
Species with culture numbers for all of the stocks used in this study. All species were obtained from the National
Drosophila Species Resource Center at Bowling Green with the exception of D. teissieri Brazzaville isofemale
line 16 which was a gift from D. Lachaise to G. Simmons. All species identifications were confirmed based on
male genitalic dissections. The following three species were originally mislabelled by the National Drosophila
Species Resource Center at Bowling Green. D. ficusphila was incorrectly labelled as D. pennae 14028–0631.0.
D. ercepeae was incorrectly labelled as D. greeni 14028–0712.0. D. greeni was incorrectly labelled as D. ercepeae
14024–0432.0. Representatives of all taxa sampled have been placed in the collections at the American Museum
of Natural History. The species, D. rajasekari (14023–0361.3) was ordered from the stock centre; however,
D. rajasekari Reddy & Krishnamurthy, 1968 and D. raychaudhurii Gupta, 1969 were made junior synonyms of
D. biarmipes Malloch, 1924 by Bock (1980). D. jambulina Parshad & Paika, 1964 has been found to be an Indian
endemic species. Collections made in Indochina (Thailand and Cambodia, e.g, D. jambulina 14028–0531.1) are
actually D. watanabei Gupta & Gupta, 1992, see Table A1.
APPENDIX B
HB CHARACTER ASSIGNMENT
By aligning the DNA or amino acid sequence, molecular systematists establish topological identity for the
primary/putative homology statement (de Pinna, 1991; Brower & Schawaroch, 1996). The total length of the hb
sequence varied from 513 bp in D. bifasciata to 456 bp in D. takahashii and D. elegans. This caused the alignment for hb gene region to be more complicated in comparison to the Adh and mt:CoII regions which had no
indels (insertions or deletions).
Alignment of hb
It was necessary to convert hb nucleotide sequence to amino acid sequence for recognition of homology (i.e.
topological identity sensuBrower & Schawaroch, 1996). Alignments were performed on hb amino acid sequences,
using the Clustal method in M E G A L I G N (D N A S TA R , version 1.02). To determine alignment ambiguous sites
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
THE MELANOGASTER SPECIES GROUP
33
Table A1.
Species
Culture/Stock
Species
Culture/Stock
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
14011–0091.0
14012–0181.0
14011–0121.0
14011–0111.0
14012–0141.0
14012–0210.0
14027–0461.0
14026–0451.0
misID 14028–0631.0
14023–0331.0
14023–0341.0
14023–0361.3
gift D. Lachaise
14021–0261.0
14022–0311.5
14022–0271.0
14022–0281.0
14022–0291.0
14024–0371.0
misID 14028–0712.0
14024–0391.0
14024–0433.0
14024–0434.0
14024–0431.0
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
14028–0471.1
14028–0491.2
14028–0481.1
14028–0501.0
14028–0511.0
14028–0521.0
14028–0586.0
misID 14024–0432.0
14028–0531.1
14028–0541.0
14028–0561.3
14028–0581.0
14028–0591.0
14028–0601.0
14028–0611.0
14028–0621.0
14028–0641.0
14028–0661.0
14028–0651.0
14028–0671.0
14028–0681.0
14028–0691.0
14028–0701.0
14028–0711.0
ambigua
bifasciata
pseudoobscura
persimilis
affinis
tolteca
elegans
eugracilis
ficusphila
lucipennis
mimetica
biarmipes
teissieri
yakuba
takahashii
lutescens
paralutea
prostipennis
ananassae
ercepeae
m. malerkotliana
pallidosa
phaeopleura
varians
auraria
barbarae
baimaii
biauraria
bicornuta
birchii
dipacantha
greeni
watanabei
kanapiae
kikkawai
lini
mayri
nikananu
orosa
parvula
punjabiensis
rufa
quadraria
seguyi
serrata
triauraria
tsacasi
vulcana
(Gatsey et al., 1994) the cost parameters varied as follows: (1) the gap length penalty was set at a value of 10;
(2) the amino acid change cost was according to the PAM250 residue weight table (Dayhoff, 1978); and (3) the
gap penalty value varied from 8 to 30.
Evaluating the alignment
Three stretches of the amino acid sequence exhibited alignment ambiguity (positions 5–20, 106–113, and 151–166
in Table B1). Future investigations with increased taxon sampling may make the third stretch (amino acids
151–166) not ambiguous and this stretch would become an excellent source of characters.
It is interesting to note that the hypervariable region predominated by Q’s and H’s at amino acid positions
30–51 was not alignment ambiguous. This may reflect the low alignment cost to switch between Q and H (a
value of 2 for a range from 0 to 22). The multiple repeats of Q’s and H’s most probably occurred by a slippage
mechanism and their putative homology statements in this region seem questionable at best. This region’s alignment, however, was conserved across all the parameters tested. Therefore, this region remained in the matrix
for analysis.
hb sequence used in phylogenetic analysis
After removal of the alignment ambiguous sites, the remaining aligned hb amino acid sequence was reconverted
to nucleotide sequence in an effort to maximize possible character information. The aligned hb nucleotide
sequence now 441 bp long was inserted back as primary data in the matrix (Table B1).
GAP
CODING
Gaps within molecular sequence have traditionally been coded as question marks. Morphological characters
coded by a question mark can be the result of one of three conditions: the character is ambiguous, inapplicable
or missing (Platnick et al., 1991). In this study gaps were neither ambiguities (due to polymorphisms) nor missing
(stretches of DNA not sequenced) but rather were inapplicable (the taxon does not have the structure [stretch
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
1
ambigua
persimilis
pseudoobscura
affinis
bifasciata
tolteca
diplacantha
watanabei
punjabiensis
greeni
kanapiae
parvula
seguyi
vulcana
nikananu
kikkawai
lini
serrata
tsacasi
orosa
auraria
triauraria
rufa
quadraria
biauraria
barbarae
birchii
mayri
bicornuta
baimaii
ananassae
phaeopleura
malerkotliana
pallidosa
varians
ercepeae
ficusphila
elegans
paralutea
prostipennis
takahashii
lutescens
lucipennis
mimetica
biarmipes
eugracilis
yakuba
teissieri
melanogaster
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SLAS
SLAS
SLAS
SLAS
SLAS
SLTS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
SVAS
*
*
GSPSPRQSPLPSP--GSPSPRQSPLPSP--GSPSPRQSPLXSP--GSPSPRQSPLPSP--GSPSPRQSPLASP--GSPSPRQSPLPSP-----SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAX
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLPA
---IPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPLPSPLAA
---SPRQSPIPSPMNP
---SPRQSPIPSPMNP
---SPRQSPIPSPMNP
---SPRQSPIPSPMNP
---SPRQSPIPSPLNP
---SPRQSPIPSPLNP
---SPRQSPIPS------SPRQSPIPS------SPRQSPIPS------SPRQSPIPS------SPRQSPIPS------SPRQSPIPS------SPRQSPIPS------SPRQSPIPS------SPRQSPIPS------SPRQSPIPS------SPRQSPIPS------SPRQSPIPS------SPRQSPIPS---¨æAmbiguousæÆ
*
*
*
*
*
*
*
97
GNHLEQYLKQQQQQ--HHQQQQLQ-----QQPMDTLCGAAMTPSPSQNDQNSLQHFDVTLHQQLLQQQQYQQHFQAA
GNHLEQYLKQQQQQ--HHQQQQLQ-----QQPMDTLCGAAMTPSPSQNDQNSLQHFDVTLQQQLLQQQQYQQHFQAA
GNHLEQYLKQQQQQ--HHQQQQLQ-----QQPMDTLCGAAMTPSPSQNDQNSLQHFDVTLQQQLLQQQQYQQHFQAA
GNHLEQYLKQQQQ----HQQQQLQ-----QQPMDTMCGAAMTPSPNQNDQNSLQHFDVTLQQQLLQQQQYQQHFQAA
GNHLEQYLKQQQQQQQHQHQQQLQ-----QQPMDTLCGAAMTPSPSQNDQNSLQHFDVTLQQQLLQQQQYQQHFQAA
GNHLEQYLKQQQHQQQ-QQQQQLQ-----QQPMDTMCGAAMTPSPSQNDQNSLQHFDVTLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-HHHQQQQQQQQ-HQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-QHHQQQQQ----HQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-QHHQQQQ-----HQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-HHHQQQQQQQQQHQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-HHHQQQQ-----HQTHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
NSQLEQFLKQQHHHQQQQQ-----HQTHQQQPMDTMC--TMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQQQHHHQQQ---QQHQSHQQQPMDXMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-HHHQQQQ---QQHQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
NSQLEQFLKQQQHH----QQQQQQHQSHQQQPMDTMC--AMTPSPSQXDQNSLQHFDATLQQQFLQQQQYQQHFQAA
SSQLEQFLKQQ-QHHQQQQ---QHHQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-HHHQQQQ---QQHQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-QHHQQQQQQQQQHQSHHQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-HHHQQQQ---QQHQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDGTLQQQLLQQQQYQQHFQAA
NSQLEQFLKQQQHHHQQQQQQQQQHQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQQHH-QQQQ---QQHQPHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQQHH-QQQQ---QQHQPHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
GSQLEQFLKQQQHH-QQQQ---QQHQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQQHH-QQQQ---QQHQPHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQQHHHQQQQ---QQHQPHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-QHHQQQ------HQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQQHHHHQQQQ----HQSQQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQHHH---QQHQEQQHQSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQ-HHQQQQQQQQQQHQSHQQQLMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
SSQLEQFLKQQQHHQQQQQHQ---HPSHQQQPMDTMC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
GNQLEQFLKQQ-HHQQQ------------QQPMDTLC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
GNQLEQFLKQQ-HHQQQ----------HQQQPMDTLC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
GNQLEQFLKQQ-QSHHQ----------QQQQPMDTLC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
GNQLEQFLKQQ-HHQQQ------------QQPMDTLC--AMTPSPSQNDQNSLQHFDATLQQQLLQQQQYQQHFQAA
ANQLEQFLKQQQHHHQQ----------QQQQPMDTLC--AMTPSPSQNDQNSLQHFDATLQQQILQQQQYQQHFQAA
GNQLEQFLKQQ-HQQHH----------HQQQPMDTLC--AMTPSPSQNDQNSLQHFDATLQQQLMQQQQYQQHFQAA
TNHLEQFLKQQQQQ-------------HQQQPMDTLC--AMTPSPSQNDQNSLQHYDANLQQQLLQQQQYQQHFQAA
TNHLEQFLKQQQ---------------HQQQPMDTLC--AMTPSPSQNDQNSLQHYDAGLQQQLLQQQQYQQHFQAA
TSHLEQFLKQQQQ--------------HQQQPMDTLC--AMTPSPSQNDQNSLQHYDASLQQQLLQQQQYQQHFQAA
TSHLEQFLKQQQQ--------------HQQQPMDTLC--AMTPSPSQNDQNSLQHYDASLQQQLLQQQQYQQHFQAA
TNHLEQFLKQQQHQ--------------QQQPMDTLC--AMTPSPSQNDQNSLQHYDASLQQQLLQQQQYQQHFQAA
TNHLEQFLKQQQ---------------HQQQPMDTLC--AMTPSPSQNDQNSLQHYDASLQQQLLQQQQYQQHFQAA
TNHLEQFLKQQHQQ--------------QQQPMDTLC--AMTPSPSQNDQNSLQHYDASLQQQLLQQQQYQQHFQAA
TNHLEQFLKQQHHQ-------------QQQQPMDTLC--AMTPSPSQNDQNSLQXYDANLQQQLLQQQQYQQHFQAA
TNHLEQFLKQQQ---------------HQQQPMDTLC--AMTPSPSQNDQNSLQHYDANLQQQLLQQQQYQQHFQAA
TNHLEQFLKQQHQQ--------------QQQPMDTLC--AMTPSPSQNDQNSLQHYDANLQQQLLQQQQYQQHFQAA
TNHLEQFLKQQQQQQ------------HQQQPMDTLC--AMTPSPSQNDQNSLQHYDASLQQQLLQQQQYQQHFQAA
TNHLEQFLKQQQQQQ------------HQQQPMDTLC--AMTPSPSQNDQNSLQHYDASLQQQLLQQQQYQQHFQAA
TNHLEQFLKQQQQQL-------------QQQPMDTLC--AMTPSPSQNDQNSLQHYDANLQQQLLQQQQYQQHFQAA
V. SCHAWAROCH
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
Species
34
Table B1a. Alignment of hb amino acid sequence for 49 taxa. The exemplar alignment is the one that resulted from using a gap penalty value of 8
98*
QQQQQQQA
QQQQQQQA
QQQQQQQA
QQQQQQQA
QQQQQQQA
QQQQQQ-A
QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ----QQQ-----
HHHHHHLG
HHHHHHLG
HHHHHHLG
HHHHHHLG
HHHHHHLG
HHHHHHLG
HHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHLHHHHHHL¨ Amb. Æ
*
*
*
*
LGGFNPLTPPGLPNPMQHFYAGNLGRPSPQPTPTATQ
LGGFNPLTPPGLPNPMQHFYAGNLGRPSPQPTPTATQ
LGGFNPLTPPGLPNPMQHFYAGNLGRPSPQPTPTATQ
LGGFNPLTPPGLPNPMQHFYAGNLGRPSPQPTPTATQ
LGGFNPLTPPGXPNPMQHFYAGNLGRPSPQPTPTATQ
LGGFNPLTPPGLPNPMQHFYAGNLGRPSPQPTPTATQ
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPTPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGSL-RPSPQPTPTT-MTGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTT-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTN-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTN-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTN-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTN-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGSL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGSL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGTL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGSL-RPSPQPTPTA-MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTAMA
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTAPS
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTSAS
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTSAS
MGGFNPLTPPGLPNPMQHFYGGSL-RPSPQPTPTSAS
MGGFNPLTPPGLPNPMQHFYGGSL-RPSPQPTPTSAS
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTSVA
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTSAS
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTSVS
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTAAA
MGGFNPLTPPGXPNPMQHFYGGNL-RPSPQPTPTSAS
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTSVS
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTSAS
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTSAS
MGGFNPLTPPGLPNPMQHFYGGNL-RPSPQPTPTSAS
*
*
*
187
VVAPTQV--------G EKLQALTPPMDVTPPKSPAKS
VVAPTQV--------G EKLQALTPPMDVTPPKSPAKS
VVAPTQV--------G EKLQALTPPMDVTPPKSPAKA
VVAPTQV--------G EKLQALTPPMDVTPPKSPAKS
VVAPTQV--------G EKLQALTPPMDVTPPKSPAKS
VVAPTQV--------G EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS DKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-GVAVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-GVAVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-GVAVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-GVAVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-GVAVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AVA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-AIA---PVAVATS EKLQALTPPMDVTPPKSPAKS
-G-TVA---TVAVATS EKLQALTPPMDVTPPKSPAKS
PSAA-----SVTSTTS EKLQALTPPMDVTPPKSPAKS
PSAA-----SVTSATS EKLQALTPPMDVTPPKSPAKS
PSAA-----SVTSATS EKLQALTPPMDVTPPKSPAKS
PSAA-----SVTSTTS EKLQALTPPMDVTPPKSPAKS
SSAA-----PVTTATS EKLQALTPPMDVTPPKSPAKS
AGTAVA---AGTAVTS EKLQALTPPMDVTPPKSPAKS
TVAS---AVPVGSATS EKLQALTPPMDVTPPKSPAKS
TIAPVAVPN-GTS--- EKLQALTPPMDVTPPKSPAKS
AVAPVALATGSSSSSS EKLQALTPPMDVTPPKSPAKS
XVAPXAXATGSSSSS- EKLQALTPPMDVTPPKSPAKS
APVAIA-----SSNNS EKLQALTPPMDVTPPKSPAKS
AVAPVAIATGSSSS-- EKLQALTPPMDVTPPKSPAKS
AVAPVAVA-NGTS--- EKLQALTPPMDVTPPKSPAKS
T-APIAVPTSSSNSSS EKLQALTPPMDVTPPKSPAKS
SVAPVAVANGGSSS-- EKLQALTPPMDVTPPKSPAKS
TVAPVAVAASSSS--- EKLQALTPPMDVTPPKSPAKS
TVAPVAVAT-GSS--- EKLQALTPPMDVTPPKSPAKS
TVAPVAVAT-GSS--- EKLQALTPPMDVTPPKSPAKS
TIAPVAVAT-GSS--- EKLQALTPPMDVTPPKSPAKS
¨æ Ambiguous æÆ
35
ambigua
persimilis
pseudoobscura
affinis
bifasciata
tolteca
diplacantha
watanabei
punjabiensis
greeni
kanapiae
parvula
seguyi
vulcana
nikananu
kikkawai
lini
serrata
tsacasi
orosa
auraria
triauraria
rufa
quadraria
biauraria
barbarae
birchii
mayri
bicornuta
baimaii
ananassae
phaeopleura
malerkotliana
pallidosa
varians
ercepeae
ficusphila
elegans
paralutea
prostipennis
takahashii
lutescens
lucipennis
mimetica
biarmipes
eugracilis
yakuba
teissieri
melanogaster
*
THE MELANOGASTER SPECIES GROUP
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
Table B1b. Continued
36
V. SCHAWAROCH
Table B2. Gaps at amino acid positions 58–59 and 138 were recoded as characters then reinserted as the primary data
in the matrix. Each of these amino acid sequences are present in the six outgroup taxa and are lost for the taxa of the
melanogaster group. Thus this deletion supports the hypothesis of ingroup monophyly. In this instance, the presence of a
gap is an informative character. There is no variation exhibited in the size of the gap; therefore, the deletion producing
the gap may have occurred only once. These data were condensed into a single unordered binary or multistate character
Species
Sequence
Coding A
Matrix B
Sequence
Coding A
Matrix B
ambigua
persimilis
pseudoobscura
affinis
bifasciata
tolteca
paralutea
prostipennis
takahashii
lutescens
lucipennis
mimetica
biarmipes
ananassae
varians
phaeopleura
greeni
malerkotliana
pallidosa
eugracilis
teissieri
yakuba
melanogaster
bicornuta
diplacantha
watanabei
punjabiensis
seguyi
vulcana
nikananu
auraria
barbarae
birchii
kikkawai
lini
quadraria
serrata
triauraria
tsacasi
baimaii
biauraria
kanapiae
mayri
orosa
parvula
rufa
ercepeae
elegans
ficusphila
GGGGCA
GGGGCA
GGGGCA
GGGGCG
GGGGCA
GGGGCG
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
000000
000000
000000
000001
000000
000001
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
111112
0
0
0
1
0
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
GGT
GGT
GGT
GGT
GGT
GGT
---------------------------------------------------------------------------------------
000
000
000
000
000
000
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37
THE MELANOGASTER SPECIES GROUP
37
of DNA] which possesses the characters). Previously, Wheeler (1993) and Danforth et al. (1999) have evaluated
gaps with respect to alignment context. Gaps at two stretches (amino acid positions 58–59 and 138) were coded
as a character because they are flanked by conserved sequence plus appear to convey grouping information. The
remaining gaps were coded in the traditional method as question marks.
COMBINATION
GAP CODING
Evaluating gap coding methods
The gaps as characters were coded in a binary form according to matrix B (Table B2). The one to one correspondence for the nucleotide characters to binary characters in ‘matrix A’ could be inflating the character information as these gaps are considered as characters and most probably occurred as single events. In contrast,
‘matrix B’ summarizes the nucleotide variation within its binary coding. Coding by either matrix produces the
same tree topology.
The effect of the coding methods (i.e. traditional all gaps as question marks and the combination gap coding
presented here) on resulting tree topologies were compared for combined data (i.e. mt:CoII +Adh +hb) and hb
data. Tree topology was unaffected by treating gaps as either all ‘missing’ or as a combination of ‘missing’ and 5th
state.
CONCLUSION
Even though these results demonstrate that the combination gap coding did not alter tree topology from the
traditional coding method (gaps as ‘missing’), combination coding reflects the information conveyed by the gaps
present in the hb sequence. Matrix B summarizes numerical coding of matrix A with the assumption that these
gaps were single events. All phylogenetic analyses in this study combination coded the gaps (both ‘missing’ and
the 5th state). PAUP was executed with the option ‘gaps as missing’, and the gaps as 5th state were coded in the
PAUP matrix using numerical values according to matrix B (Table B2).
© 2002 The Linnean Society of London, Biological Journal of the Linnean Society, 2002, 76, 21–37