MOLECULARPHYLOGENETICSAND
Molecular Phylogenetics and Evolution 30 (2004) 236–242
EVOLUTION
www.elsevier.com/locate/ympev
Brief communication
The advantages of the ITS2 region of the nuclear rDNAcistron for analysis of phylogenetic relationships of insects:
a Drosophila example
Irene Young and Annette W. Coleman*
Division of Biology and Medicine, Brown University, Providence, RI 02912, USA
Received 18 October 2002; revised 9 April 2003
Abstract
We examined the utility for phylogenetic reconstruction of the second internal transcribed spacer (ITS2), lying between the
nuclear 5.8S gene and the gene for large subunit ribosomal RNA, using sequences of Ceratitis, Bactrocera, Musca, and Drosophila.
We aligned and analyzed 13 sequences from GenBank and 11 new sequences from diverse species of Drosophila. Derivation of the
secondary structure of the ITS2, the RNA transcript folding pattern required for transcript processing into functional RNA units,
revealed the facets of sequence conservation common to all the sequences, that then allowed alignment of all the genera. The re-
sultant tree, though including only a sparse representation of the enormous Drosophila diversity, conforms generally with the
consensus of all prior phylogenetic reconstructions, using eight other nuclear and mitochondrial gene regions; where species rep-
resentation is greater, as in the melanogaster subgroup of the Sophophora subgenus representatives, it conforms exactly. The par-
adigm ITS2 secondary structure presented can now be used to assess the genus more thoroughly, since its base pairing pattern makes
alignment of sequences obvious. In addition, it shows that these insects share the ITS2 secondary structure characteristics of the
other major animal groups as well as the green line of eukaryote evolution. The relatively short (<400 bp) ITS2 region seems ideal
for reconstructing evolutionary relationships at the levels of species, genera, and perhaps even higher.
� 2003 Elsevier Science (USA). All rights reserved.
1. Introduction
Selection of DNA regions suitable for phylogenetic
comparisons among species and genera, even between
families, is always a challenge (Brower and DeSalle,
1994). One wants a region that faithfully reflects geneticrelationships and that can be selectively multiplied by
PCR with ease. Economic and time constraints en-
courage use of a relatively short region with high in-
formation content.
The nuclear ribosomal repeat cistrons have been
widely used for phylogenetic studies of protistan, plant,
and animal species. The nuclear RNA gene sequences
have already proven valuable for higher taxonomiclevels of insects (Pelandakis and Soliqnac, 1993). A re-
gion of these repeats more suitable for genus and species
* Corresponding author. Fax: 1-401-863-1182.
E-mail address: [email protected] (A.W. Coleman).
1055-7903/$ - see front matter � 2003 Elsevier Science (USA). All rights res
doi:10.1016/S1055-7903(03)00178-7
comparisons is the second Internal Transcribed Spacer
region (ITS2), the one that lies between the 5.8S and
large subunit RNA genes. It is typically 200–400 bp in
length, easily amplified by PCR from even miniscule
amount of DNA, and easy to sequence.
Since the ultimate product of the ITS2 regions is anRNA, not a protein with its innate triplet punctuation,
the proper alignment of ITS2 sequences at or above the
genus level has proved challenging in the absence of
other information, for although some subregions are
very highly conserved, others seem free to vary at ran-
dom (Schl€ootterer et al., 1994). Without proper align-
ment, there is little sense in applying programs analyzing
phylogenetic relationships. Even when analysis is limitedto the subregions of unambiguous alignment, significant
information content at the species level is lost. The so-
lution to the alignment problem for ITS regions is the
recognition of the secondary structure formed by the
folding of the primary RNA transcript, a secondary
structure necessary for the ‘‘processing’’ in the nucleolus
erved.
I. Young, A.W. Coleman / Molecular Phylogenetics and Evolution 30 (2004) 236–242 237
of the long primary transcript of the ribosomal cistron(Venema and Tollervey, 1999). The products of pro-
cessing are the small and large subunits of RNA and the
5.8S RNA of the ribosome. During processing, the ITS
regions are degraded to nucleotides.
Although some of the aspects of the ITS2 secondary
structure necessary for processing are now recognized,
the actual folding pattern has largely been established by
the same procedures used to determine the secondarystructure of the RNA components of the ribosomes
themselves. This method involves comparison of po-
tential RNA folding patterns of closely related taxa to
determine which single example is common to all and
supported by compensatory nucleotide substitutions
that always preserve the pairing necessary for this sec-
ondary structure.
One of the first phylogenetic explorations of ITS2using RNA secondary structure as a guide to alignment
was the study of eight Drosophila species by Schl€oottereret al. (1994). Their proposed ITS2 secondary structure
was both incomplete and inapplicable, except for one
hairpin loop, to their outgroup, Musca. Since 1994, the
ITS2 secondary structure has been determined for a
wide variety of eukaryote groups, both plant and ani-
mal, and most recently, with the refolding of the originalyeast model (Joseph et al., 1999), it has become clear
that all these eukaryote groups share the same overall
secondary structure (Coleman and Vacquier, 2002; Mai
and Coleman, 1997; Michot et al., 1999). These devel-
opments open a much broader vista for phylogenetic
application of ITS2.
Since the universal ITS2 secondary structure now
recognized (Coleman, 2003) differs from that originallyproposed by Schl€ootterer et al. (1994), we decided to re-
visit the drosophilid case. Specifically, we wished to
determine if there is indeed a single secondary structure
that characterizes all drosophilids, and if it conforms
with the general eukaryote model. If so, it would make
ITS2 valuable once again for phylogenetic analyses in
this group, and if alignment guided by secondary
structure led to success in aligning even more distantlyrelated insects, a much wider applicability in insect
evolutionary studies.
2. Materials and methods
Data on the species of Drosophila sequenced for
this study are listed in Fig. 2, along with ITS2 se-quences available from GenBank that were also used.
To obtain template DNA, flies were squashed and the
DNA extracted using the protocol for single fly DNA
extraction given as ‘‘1995’’ on the FlyBase website
(http://flybase.bio.indiana.edu). Two microliters of this
DNA extract was added to a 50 ll mixture of buffer,
primers, deoxynucleotides, and MgCl2, as described in
the Taq polymerase protocol of Promega (Madison,WI). After the mixture had first reached 95 �C, 1–
1.5U of Taq polymerase were added. The thermocy-
cler profile was that of Schl€ootterer et al. (1994), and
the forward and reverse primers we used initially are
given in that paper. Subsequently, an internal forward
primer representing a sequence in the 5.8S region
specific to drosophilids (‘‘D5.8S for’’¼ 50-AGAACG
AGCAAACTGTGC-30) was substituted for the origi-nal forward primer, which is near the 30 end of the
SSU RNA gene, and a shorter PCR program was
used (see Coleman et al., 2001).
PCR products were purified from agarose gels by use
of the Qiaquick Gel Extraction kit (Qiagen, Valencia,
CA) and sequenced in both directions, using the ABI
Prism 377 DNA sequencer and Big Dye Chemistry
protocol (Applied Biosystems, Foster City, CA) and thesame primers as for PCR. In two cases, PCR products
were also subcloned into pGem T-vector (Promega,
Madison, WI) for comparison with sequence generated
from direct PCR sequencing.
Sequences were edited and aligned using MacVector
software (Kodak, International Biotechnologies, New
Haven, CT), and the termini of ITS2 conform to those
used in Schl€ootterer et al. (1994). Optimal alignment wasgreatly aided by knowledge of the secondary structure of
the ITS2 region of the primary transcript. This was es-
tablished by submission of the primary sequence of all
the species to the RNA folding website supporting mfold
version 3.1 (available from http://bioinfo.math.rpi.edu/
~mfold/rna/form1.cgi) using the default parameters for
folding (Zuker et al., 1999). Comparisons among the
results for the various species revealed the folding pat-tern common to them all, which in turn established the
regions of relatively conserved primary structure and
hence homology for alignment. Thus, pairing positions
on the 50 side of a hairpin were aligned, and their cor-
responding pairing partners on the 30 side were likewise
aligned for each sequence.
Aligned sequences were subjected to phylogenetic
analysis using PAUP* version 4.0b10 (Swofford, 2002).All nucleotides were weighted equally, and gaps were
treated as missing data. Both parsimony (branch and
bound analysis for tree construction) and distance
(matrix generated by the Kimura two parameter algo-
rithm and tree building by neighbor joining) methods,
with other parameters left at the default position, were
used to derive trees, and variations were tried that in-
cluded all nucleotides, only those present in regions I, II,and III, and only those in regions I and II. Coding gaps
as ‘‘fifth nucleotide’’ did not add any detail to the trees.
The sequence of Musca served as outgroup to the dro-
sophilids. Bootstrap support was evaluated with 100
(parsimony) or 500 (neighbor joining) iterations. New
sequences have been deposited in GenBank and the
alignment is available upon request.
238 I. Young, A.W. Coleman / Molecular Phylogenetics and Evolution 30 (2004) 236–242
3. Results
With either the Schl€ootterer et al. (1994) primer pair or
our ‘‘D5.8S for’’ plus the Schl€ootterer reverse primer,
typically only a single band was found when the PCR
products were run on an agarose gel. Its identity as the
nuclear ribosomal region was confirmed by comparison
of the very conservative 5.8S sequence with that of
previously sequenced insect examples. The length of theDrosophila ITS2, as determined by sequencing, was 320–
429 bp and the percent GC content ranged from 18 to
26% for the full-length sequences.
3.1. RNA transcript secondary structure
The putative secondary structure of the ITS2 RNA
transcript region of Drosophila melanogaster is shown inFig. 1, and serves for all the organisms sequenced, in-
cludingMusca. It is fundamentally similar to that of other
eukaryotes in having: (a) four helix loop regions, of which
(b) the helix in region II is highly conserved and bears a
pyrimidine mismatch within the basal seven nucleotide
pairings, and (c) helix III displays on its 50 side, near theterminus, the single most conserved region of primary
sequence among all the species.A stretch of 10 nucleotidesthere (marked in Fig. 1) is identical among all the Dro-
sophila sequences, as well asMusca,Bactrocera,Ceratitis,
and even Glossina, the tse-tse fly (Chen et al., 1999). This
region is part of a 25 nucleotide sequence essentially
identical among all Drosophila species (also marked in
Fig. 1). There are two sites of variation: Drosophila cru-
cigera, the Hawaiian species, has UC rather than CU in
the nucleotide bulge on the 50 side; and for the A–Upairingmarked by the arrow, all the subgenusSophophora
have the A–U pairing illustrated except Drosophila fi-
cusphila (G–U), Drosophila sturtevant and Drosophila
willistoni (G–C), and Drosophila pseudoobscura (A–C);
the remaining Drosophila species have G–C, except D.
crucigera (A–U). In each case, the nucleotide substitution
is such that pairing potential is maintained.
The helix in region I is also remarkably similaramong the species. In region IV, traditionally the most
variable region of ITS2, D. melanogaster and its five
closest relatives clearly have the potential for two loops
(IVA and IVB in Fig. 1), while the other species and
genera probably have but one, perhaps Y-shaped, the
more standard situation in other organisms. The proof
of secondary structure in region IV would require se-
quencing additional related species for comparisons.One characteristic of the drosophilid secondary struc-
ture that is not often found among other eukaryotes is
the short helix labeled IIa that appears between helix II
and helix III. This is totally absent in all the non-dro-
sophilids examined.
The ‘‘proof’’ of putative secondary structure in RNA
has been the presence of CBCs (Compensatory Base
Changes) as defined in Gutell and Larsen (1994). A CBCis a pairing position in a helix where the sequences of
two related organisms differ at both positions yet retain
the pairing potential (for example, see Coleman, 2003).
Among just the drosophilids used here, there are six
CBCs among the basal nine pairings of helix I; the basal
10 pairings of helix II are identical and the 11th pairing
has a CBC, as does the 12th; and in the relatively con-
served region of helix III marked in Fig. 1 with an ar-row, there is the one position where CBCs occur, as
already noted by Schl€ootterer et al. (1994). Additional
CBCs and numerous hemi-CBCs, pairings where only
one the of the nucleotides is altered appropriately (e.g.,
G–C becomes G–U), are found in the less conserved
regions of helix.
3.2. Phylogenetic analysis
The total ITS2 alignment of Musca and 18 species of
Drosophila evaluated with PAUP* had 597 positions of
which 195 were parsimony informative. The single most
parsimonious tree (Fig. 2) has superimposed on it
bootstrap values obtained from both parsimony and
distance analyses. There were no conflicts among all
methods of tree building as to which clades emerged,only in the level of support for various of these clades.
The tree topology was unchanged, whether using the
entire ITS2 length or only selected regions encompassing
the more conserved helices.
Fig. 2 (inset A) presents a distance tree including
Ceratitis, Bactrocera, Musca, and all the Drosophila
species used. It reaffirms the monophyletic nature of
these drosophilids with respect to the outgroups, but forfurther phylogenetic analysis, the Ceratitis and Bactro-
cera sequences were omitted because they added no
clarity to the trees, only homoplasy. The Drosophila
serrata sequence was also omitted because the sequence
was incomplete (only regions I, II, and 50 of III), but it isclearly most similar to Drosophila takahashi.
4. Discussion
The selection of Drosophila species used here was
chosen both to overlap with previous studies and to
encompass a sufficient range of the diversity within this
large genus to derive and analyze the common ITS2
transcript folding. Both in our experience, and that of
Schl€ootterer, polymorphisms for the entire ITS1-5.8S-ITS2 are minimal (0–0.05%). We failed to find poly-
morphism within an individual, and our sequences of
Drosophila virilis and D. pseudoobscura were identical to
those already in GenBank, suggesting that concerted
evolution (Dover, 1982) has homogenized ITS2 of the
numerous ribosomal repeats sufficiently for its use in
phylogenetic analysis.
Fig. 1. Secondary structure diagram of the primary RNA transcript of Drosophila melanogaster ITS2. The four major folding domains of ITS2 are
designated with roman numerals. The characteristic pyrimidine mismatch of helix II is indicated by an arrowhead. In the region of highest primary
sequence conservation, on the 50 side of helix III, the 10-nucleotide primary sequence common to Drosophila, Musca, Bactrocera, Ceratitis, and
Glossina is circled, while the 25-nucleotides very highly conserved among the drosophilids are bracketed. Arrows indicate the nucleotide pairing in
this conserved portion of the helix that shows CBCs within the genus Drosophila.
I. Young, A.W. Coleman / Molecular Phylogenetics and Evolution 30 (2004) 236–242 239
Approximately 115 nucleotide positions in ITS2 are
relatively conserved (Mai and Coleman, 1997). In
agreement with this, Schl€ootterer et al. (1994) estimated
that 40% of the total ITS is constrained by selection in
Drosophila, and the remainder they found to be un-
constrained positions evolving with a rate that is close
to the neutral rate in this species group. Thus ITS2
combines information most germane to species and
subspecies levels with that useful for comparing even
genera and higher levels.
240 I. Young, A.W. Coleman / Molecular Phylogenetics and Evolution 30 (2004) 236–242
b
Fig. 2. Single most parsimonious tree found by branch and bound analysis of Drosophila species, using all nucleotide positions of ITS2 and Musca as
outgroup. Subgenera designations of drosophilids are shown. Tree length¼ 726 steps, consistency index¼ 0.705, and homoplasy index¼ 0.299. Step
length is shown below each line, and bootstrap values greater than 50% above each line, the first from parsimony and the second from distance
analysis (Kimura two parameter, neighbor joining) of the same data. Inset A: Distance tree, with bootstrap values, using the Drosophila and Musca
sequences, plus Ceratitis and Bactrocera, rooted with the Musca sequence. GenBank sequences used are D. virilis Z28415, D. pseudoobscura Z28460,
Drosophila yakuba Z28416, Drosophila orena Z28549, Drosophila simulans Z28413, Drosophila sechellia Z28412, Drosophila mauritiana Z28538, D.
melanogaster M21017, Musca domistica Z28417, Bactrocera dorsalis AF276516, Bactrocera pyrifoliae AF332590, Bactrocera cucumis AF276515,
Ceratitis capitata AF189691. New sequences, with their Tucson Stockcenter Nos. (//stockcenter.arl.arizona.edu), are D. busckii (#13000-0081.0)
AF551741, Drosophila flavomontana (#15010-0981.0) AF551742, D. crucigera (#15287-2531.0) AF551743, D. takahasii (#14022-0311.0) AF551744,
D. ananassae (#14024-0371.0) AF551750, D. ficusphila (#14025-0441.1) AF551747, Drosophila eugracilis (#14026-0451.1) AF551745, Drosophila
elegans (#14027-0461.0) AF551746, D. serrata (#14028-0681.0) AY175379, D. sturtevanti (#14043-0871.1) AF551748, D. willistoni (#14030-0811.0)
AF551749.
I. Young, A.W. Coleman / Molecular Phylogenetics and Evolution 30 (2004) 236–242 241
Whereas the earlier study that included ITS2(Schl€ootterer et al., 1994) failed to obtain a secondary
structure applicable to all the species used, and hence
missed such aspects as recognition of helix I and its
homologous helix in Musca, our secondary structure
reveals all four major regions and their parallels with
ITS2 of other eukaryotes. This permits alignment of not
only the drosophilids, but also Musca, Bactrocera,
Ceratitis, and Glossina, at least for regions I, II, and III,with great confidence. A cursory examination of the
ITS2 sequences of Chironomus and Glyptotendipes
available from GenBank shows that the first three heli-
ces are easily recognizable, as is the most conserved re-
gion of primary structure, that on the 50 side of helix III.
Thus ITS2 alignments might be made of all the Diptera
for regions I, II, and III.
4.1. Phylogenetic comparisons with other data sets
It was not our purpose to produce a detailed phylo-
genetic analysis of the drosophilids, a project that could
now be carried out using many of the species we have
omitted. The only previous evaluation using ITS was
that of Schl€ootterer et al. (1994) for eight Drosophila
species. In their ITS tree, the arrangement is identical tothat in Fig. 2.
Previous analyses of subsets of Drosophila species
overlapping with those included here made use of se-
quences from the large subunit nuclear ribosomal DNA
gene (Pelandakis and Soliqnac, 1993), mitochondrial
DNA (e.g., DeSalle, 1992), and nuclear genes Adh, Sod,
Gpdh (Katoh et al., 2000; Kwiatowsky and Ayala, 1999;
Russo et al., 1995) and Ddc and amd (Tatarenkov et al.,2001). Further analyses combining data from these
sources and additional species include O�Grady and
Kidwell (2002) and Remsen and O�Grady (2002). Within
the Sophophora subgenus the only notable difference
from these previous studies is the position of D. pseud-
oobscura basal to the Drosophila sturtevanti–Drosophila
willistoni–Drosophila ananassae clade. For the moment,
we attribute this to the absence in our study of the ad-ditional species expected to clarify the basal regions of
the tree. Bettencourt and Feder (2001) summarize the
distribution of species having duplicated heat shockprotein sequences (hsp70), representing a major event in
the Sophophora phylogenetic history. It is particularly
interesting to note that exactly that branch, the clade of
D. ficusphila and more terminal species, is supported by
very high bootstrap values in Fig. 2.
The tree in Fig. 2 (inset A), as in previous studies
limited to these representatives, finds the genus Dro-
sophila monophyletic. The subgenus Drosophila includesthe Hawaiian species, as also found with previous mo-
lecular studies, and the subgenus Sophophora includes
the D. willistoni subspecies clade. Not all of the previous
analyses included the interesting species, Drosophila
busckii, generally treated as belonging to a third sub-
genus, Dorsilopha. A trichotomy directly into the three
subgenera is found in Fig. 2 (inset A).
In summary, the ITS2 evaluation alone, guided bytranscript secondary structure for alignment, does not
conflict with significant support from expectations
from previous studies, and reproduces in remarkable
detail, considering the paucity of included species, the
consensus tree supported by eight other nuclear and
mitochondrial DNA regions. What our complement of
Drosophila ITS2 sequences has shown is that phylog-
eny as analyzed by ITS2 is at least as informative asthat obtained by using any other DNA locus. Where
sufficient species representation is present, the highly
supported clades are the same. The insects, ecdysids,
appear to share the same ITS2 characteristics as do
the other protostome and the deuterostome animal
groups, and higher fungi and plants as well. In 2001,
Tatarenkov et al. suggested in their review of molec-
ular data that ‘‘no single gene has yet produced anunequivocal phylogeny of the Drosophilidae’’; perhaps
the ITS2 region of DNA will contribute very signifi-
cantly to that goal.
Acknowledgments
The authors gratefully acknowledge the kind gift ofdrosophilids from Dr. Kristi Wharton and Dr. David
Rand, Brown University.
242 I. Young, A.W. Coleman / Molecular Phylogenetics and Evolution 30 (2004) 236–242
References
Bettencourt, R.R., Feder, M.E., 2001. Hsp70 duplication in the
Drosophila melanogaster species group: how and when did two
become five? Mol. Biol. Evol. 18, 1272–1282.
Brower, A.V.Z., DeSalle, R., 1994. Practical and theoretical
considerations for choice of a DNA sequence region in insect
molecular systematics, with a short review of published studies
using nuclear gene regions. Ann. Entomol. Soc. Am. 87, 702–
716.
Chen, X., Li, S., Aksoy, S., 1999. Concordant evolution of a symbiont
with its host insect species: molecular phylogeny of genus Glossina
and its bacteriome-associated endosymbiont, Wigglesworthia glos-
sinidia. J. Mol. Evol. 48, 49–58.
Coleman, A.W., 2003. ITS2 is a double-edged tool for eukaryote
evolutionary comparisons. Trends Genet. 19, (in press).
Coleman, A.W., Vacquier, V.D., 2002. Exploring the phylogenetic
utility of ITS sequences for animals: a test case for abalone
(Haliotis). J. Mol. Evol. 54, 246–257.
Coleman, A.W., Jaenicke, L., Starr, R.C., 2001. Genetics and sexual
behavior of the pheromone producer, Chlamydomonas allensworthii
(Chlorophyceae). J. Psychol. 37, 1–5.
DeSalle, R., 1992. The phylogenetic relationships of flies in the family
Drosophilidae deduced from mtDNA sequences. Mol. Phylogenet.
Evol. 1, 31–40.
Dover, G., 1982. Molecular drive: a cohesive mode of species
evolution. Nature 299, 111–117.
Gutell, R.R., Larsen, N., Woese, C.R., 1994. Lessons from an evolving
rRNA: 16S and 23S rRNA structures from a comparative
perspective. Microbiol. Rev. 58, 10–26.
Joseph, N., Krauskopf, E., Vera, M.I., Michot, B., 1999. Ribosomal
internal transcribed spacer 2 (ITS2) exhibits a common core of
secondary structure in vertebrates and yeast. Nucleic Acids Res. 27,
4533–4540.
Katoh, T., Tamura, K., Aotsuka, T., 2000. Phylogenetic position of
the subgenus Lordiphosa of the genus Drosophila (Diptera:
Drosophilidae) inferred from alcohol dehydrogenase (Adh) gene
sequences. J. Mol. Evol. 51, 122–130.
Kwiatowsky, J., Ayala, F.J., 1999. Phylogeny of Drosophila and
related genera: conflict between molecular and anatomical analy-
ses. Mol. Phylogenet. Evol. 13, 319–328.
Mai, J.C., Coleman, A.W., 1997. The internal transcribed spacer 2
exhibits a common secondary structure in green algae and
flowering plants. J. Mol. Evol. 44, 258–271.
Michot, B., Joseph, N., Mazan, S., Bachellerie, J.P., 1999. Evolution-
arily conserved structural features in the ITS2 of mammalian pre-
rRNAs and potential interactions with the snoRNA U8 detected
by comparative analysis of new mouse sequences. Nucleic Acids
Res. 27, 2271–2282.
O�Grady, P.M., Kidwell, M.G., 2002. Phylogeny of the subgenus
Sophophora (Diptera: Drosophilidae) based on combined analysis
of nuclear and mitochondrial sequences. Mol. Phylogenet. Evol.
22, 442–453.
Pelandakis, M., Soliqnac, M., 1993. Molecular phylogeny of Drosoph-
ila based on ribosomal RNA sequences. J. Mol. Evol. 37, 525–543.
Remsen, J., O�Grady, P., 2002. Phylogeny of Drosophilinae (Dip-
tera:Drosophilidae), with comments on combined analysis and
character support. Mol. Phylogenet. Evol. 24, 249–264.
Russo, C.A.M., Takezaki, N., Nei, M., 1995. Molecular phylogeny
and divergence times of Drosophilid species. Mol. Biol. Evol. 12,
391–404.
Schl€ootterer, C., Hauser, M.T., Haeseler, A.V., Tautz, D., 1994.
Comparative evolutionary analysis of rDNA ITS regions in
Drosophila. Mol. Biol. Evol. 11, 513–522.
Swofford,D.L., 2002. PAUP*. Phylogenetic analysis using parsimony (*
andothermethods).Version4. SinauerAssociates, Sunderland,MA.
Tatarenkov, A., Zurovcova, M., Ayala, F.J., 2001. Ddc and amd
sequences resolve phylogenetic relationships of Drosophila. Mol.
Phylogenet. Evol. 20, 321–325.
Venema, J., Tollervey, D., 1999. Ribosome synthesis in Saccharomyces
cerevisiae. Annu. Rev. Genet. 33, 261–311.
Zuker, M., Mathews, D.H., Turner, D.H., 1999. Algorithms and
thermodynamics for RNA secondary structure prediction: a
practical guide. In: Barciszewski, J., Clark, B.F.C. (Eds.), RNA
Biochemistry and Biotechnology. NATO ASI Series. Kluwer
Academic Publishers, Hingham, MA, pp. 11–43.