Homology-based annotation of non-coding RNAs in the genomes of Schistosoma mansoni and Schistosoma japonicum

BioMed Central

BMC Genomics

Open AcceResearch articleHomology-based annotation of non-coding RNAs in the genomes of Schistosoma mansoni and Schistosoma japonicumClaudia S Copeland12 Manja Marz1 Dominic Rose1 Jana Hertel1 Paul J Brindley2 Clara Bermudez Santana18 Stephanie Kehr1 Camille Stephan-Otto Attolini3 and Peter F Stadler14567

Address 1Bioinformatics Group Department of Computer Science and Interdisciplinary Center for Bioinformatics University of Leipzig Haumlrtelstraszlige 16-18 D-04107 Leipzig Germany 2Department of Microbiology Immunology amp Tropical Medicine George Washington University Medical Center 2300 I Street NW Washington DC 20037 USA 3Memorial Sloan-Kettering Cancer Center Computational Biology Department 1275 York Avenue Box 460 New York NY 10065 USA 4Max Planck Institute for Mathematics in the Sciences Inselstrasse 22 D-04103 Leipzig Germany 5Fraunhofer Institute for Cell Therapy and Immunology Perlickstraszlige 1 D-04103 Leipzig Germany 6Santa Fe Institute 1399 Hyde Park Rd Santa Fe NM 87501 USA 7Institute for Theoretical Chemistry University of Vienna Waumlhringerstraszlige 17 A-1090 Wien Austria and 8Department of Biology National University of Colombia Carrera 45 No 26-85 Bogotaacute DC Colombia

Email Claudia S Copeland - cclaudiabioinfuni-leipzigde Manja Marz - manjabioinfuni-leipzigde Dominic Rose - dominicbioinfuni-leipzigde Jana Hertel - janabioinfuni-leipzigde Paul J Brindley - mtmpjbgwumcedu Clara Bermudez Santana - clarabioinfuni-leipzigde Stephanie Kehr - steffibioinfuni-leipzigde Camille Stephan-Otto Attolini - camillebioinfuni-leipzigde Peter F Stadler - studlabioinfuni-leipzigde

Corresponding author

AbstractBackground Schistosomes are trematode parasites of the phylum Platyhelminthes They are consideredthe most important of the human helminth parasites in terms of morbidity and mortality Draft genomesequences are now available for Schistosoma mansoni and Schistosoma japonicum Non-coding RNA(ncRNA) plays a crucial role in gene expression regulation cellular function and defense homeostasis andpathogenesis The genome-wide annotation of ncRNAs is a non-trivial task unless well-annotated genomesof closely related species are already available

Results A homology search for structured ncRNA in the genome of S mansoni resulted in 23 types ofncRNAs with conserved primary and secondary structure Among these we identified rRNA snRNA SLRNA SRP tRNAs and RNase P and also possibly MRP and 7SK RNAs In addition we confirmed fivemiRNAs that have recently been reported in S japonicum and found two additional homologs of knownmiRNAs The tRNA complement of S mansoni is comparable to that of the free-living planarian Schmidteamediterranea although for some amino acids differences of more than a factor of two are observed LeuSer and His are overrepresented while Cys Meth and Ile are underrepresented in S mansoni On theother hand the number of tRNAs in the genome of S japonicum is reduced by more than a factor of fourBoth schistosomes have a complete set of minor spliceosomal snRNAs Several ncRNAs that are expectedto exist in the S mansoni genome were not found among them the telomerase RNA vault RNAs and YRNAs

Conclusion The ncRNA sequences and structures presented here represent the most complete datasetof ncRNA from any lophotrochozoan reported so far This data set provides an important reference forfurther analysis of the genomes of schistosomes and indeed eukaryotic genomes at large

Published 8 October 2009

BMC Genomics 2009 10464 doi1011861471-2164-10-464

Received 27 May 2009Accepted 8 October 2009

This article is available from httpwwwbiomedcentralcom1471-216410464

copy 2009 Copeland et al licensee BioMed Central Ltd This is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby20) which permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

of 13(page number not for citation purposes)

BMC Genomics 2009 10464 httpwwwbiomedcentralcom1471-216410464

BackgroundNon-coding RNA (ncRNA) plays a crucial role in geneexpression regulation cellular function and defense anddisease Indeed in higher eukaryotes most of thegenomic DNA sequence encodes non-protein-codingtranscripts [1] In contrast to protein-coding mRNAsncRNAs do not form a homogeneous class The best-char-acterized subclasses form stable basepairing patterns (sec-ondary structures) that are crucial for their function Thisgroup includes the well-known tRNAs catalytically activeRNAs such as rRNA snRNAs RNase P RNA and otherribozymes and regulatory RNAs such as microRNAs andspliceosomal RNAs that direct protein complexes to spe-cific RNA targets Much less is known about long mRNA-like ncRNAs which are typically poorly conserved at thelevel of both sequence and structure

Most non-vertebrate genome projects have put littleemphasis on a comprehensive annotation of ncRNAsIndeed most non-coding RNAs with the notable excep-tion of tRNAs and rRNAs are difficult or impossible todetect with BLAST in phylogenetically distant organismsHence ncRNA annotation is not part of generic genomeannotation pipelines Dedicated computational searchesfor particular ncRNAs for example RNase P and MRP[23] 7SK RNAs [45] or telomerase RNA [67] are verita-ble research projects in their own right Despite bestefforts ncRNAs across the animal phylogeny remain to alarge extent uncharted territory

The main difficulty with ncRNA annotation is poorsequence conservation and indel patterns that often corre-spond to large additional expansion domains In manycases the secondary structure is much better conservedthan the primary sequence providing a means of confirm-ing candidate ncRNAs even in cases where sequence con-servation is confined to a few characteristic motifsSecondary structure conservation can also be utilized todetect homologs of some ncRNAs based on characteristiccombinations of sequence and structure motifs using spe-cial software tools designed for this purpose

In [8] we described a protocol for a more detailed homol-ogy-based ncRNA annotation than what can be achievedwith currently available automatic pipelines Here weapply this scheme to the genome of S mansoni and bycomparison with the newly sequenced S japonicumgenome identify ncRNAs in both of these clinicallyimportant schistosomes

Schistosomes belong to an early-diverging group withinthe Digenea but are clearly themselves highly derived [9-11] The flatworms are a long-branch group suggestingrapid mutation rates (see [12])

Schistosome genomes are comparatively large estimatedto be over 350 megabase pairs and perhaps as high as 400megabase pairs for the haploid genome of S mansoni andS japonicum [13-15] The other major schistosome speciesparasitizing humans probably have a genome of similarsize based on the similarity in appearance of their karyo-types [16] These large sizes may be characteristic of platy-helminth genomes in general the genome of Schmidteamediterranea is even larger with the current genomesequencing project reporting a size of ~480 million basepairs [17]httpgenomewustledugenomesviewschmidtea_mediterranea

Genome sequencing of the seven autosomes and the pairof sex chromosomes of S mansoni with about 8times coveragehas lead to a genome assembly comprising 5745 scaffolds(gt 2 kb) covering 363 Mb [131418] Similarly shotgunsequencing of S japonicum with coverage of 54times decoded397 Mb of sequence [15] These form about 25000 scaf-folds Albeit both genome projects did not lead to com-plete finished genomes we therefore know at least 90-95 of the genomic DNA sequences of S japonicum andS mansoni respectively

The protein-coding portion of the Schistosoma genomeshave received much attention in recent years Publishedwork includes transcriptome databases for both S japoni-cum [19] and S mansoni [20] microarray-based expres-sion analysis [21] characterization of promoters [2223]and physical mapping and annotation of protein-codinggenes from both the S mansoni and S japonicum genomeprojects [18] Recently a systematic annotation of pro-tein-coding genes in S japonicum was reported [24] Incontrast to other better-understood parasites such asPlasmodium [25] however not much is known about thenon-coding RNA complement of schistosomes Only thespliced leader RNA (SL RNA) of S mansoni [26] the ham-mer-head ribozymes encoded by the SINE-like retrotrans-posons Sm-α and Sj-α [2728] and secondary structureelements in the LTR retrotransposon Boudicca [29] havereceived closer attention Ribosomal RNA sequences havebeen available mostly for phylogenetic purposes [30] andtRNAs have been studied to a limited degree [31]

The wealth of available ESTs in principle provides a val-uable resource for ncRNA detection Since mostly poly-AESTs have been generated it is not surprising that mostESTs have been attributed to protein-coding genes [32]The large evolutionary distance with 55 of the geneswithout homologs outside the genus [1318] makes ithard or even impossible to reliably distinguish ESTs ofputative mRNA-like ncRNAs from non-coding portions ofprotein-coding transcripts

In this contribution we therefore focus on a comprehen-sive overview of the evolutionary conserved non-codingRNAs in the genomes of S mansoni and S japonicum Wediscuss representatives of 23 types of ncRNAs that weredetected based on both sequence and secondary structurehomology

Results and discussionStructure and homology-based searches of the schisto-some genomes revealed ncRNAs from 23 different RNAcategories Table 1 lists these functional ncRNA categoriesthe number of predicted genes in each category and refer-ences associated with each RNA type Supplementaryfasta files containing the ncRNA genes bed files withthe genome annotation and stockholm-format align-ment files can be accessed at httpwwwbioinfuni-leipzigdePublicationsSUPPLEMENTS08-014

Transfer RNAsCandidate tRNAs were predicted with tRNAscan-SE inthe genomes of S mansoni S japonicum and S mediterra-nea (a free-living platyhelminth used for comparison)After removal of transposable element sequences (seebelow) tRNAscan-SE predicted a total of 713 tRNAs forS mansoni and 739 for S mediterranea while 154 tRNAswere found in the S japonicum sequences These includedtRNAs encoding the standard 20 amino acids of the tradi-tional genetic code selenocysteine encoding tRNAs(tRNAsec) [33] and possible suppressor tRNAs [34] in allthree genomes The tRNAsec from schistosomes has beencharacterized and is similar in both size and structure totRNAsec from other eukaryotes [35]

The tRNA complements of the three platyhelminthgenomes are compared in detail in Figure 1 The amino

Table 1 Summary of homology-based RNA annotations from the sequenced genomes of S mansoni and S japonicum

RNA class Functional Category S man S jap Related reference(s)

7SK Transcription regulation (1) 0 This study

Hammerhead ribozymes Self-cleaving gt 38 000 gt 5 000 [27]

miRNA Translation control 8 7 [109] this study

potassium channel motif RNA editing 9 3 [65]

RNase MRP Mitochondrial replication rRNA processing (1) (1) This study

RNase P tRNA processing 1 1 This study

rRNA-operon Polypeptide synthesis 80-105 50-280 [39] this study5S rRNA Polypeptide synthesis 21 1-13 This study

SL RNA Trans-splicing 6-48 1-9 [26] this study

SnoRNA U3 Nucleolar rRNA processing 1 1 This study

SRP Protein transportation 12 4+1 This study

tRNA Polypeptide synthesis 663 154 This study

U1 Splicing 3-34 2-6 [44] this studyU2 Splicing 3-15 1-63 [44] this studyU4 Splicing 1-19 1-6 [44] this studyU5 Splicing 2-9 1-24 [44] this studyU6 Splicing 9-55 2-12 [44] this studyU11 Splicing 1 1 This studyU12 Splicing 1-2 0-1 [44] this studyU4atac Splicing 1 1 This studyU6atac Splicing 1 1 This study

U7 Histone maturation 0 (2) This study

Where are range of numbers is given it remains uncertain whether multiple copies in the genomic DNA are true copies of the gene or assembly artifacts

Comparison of the tRNA complement of Schistosoma mansoni Schistosoma japonicum and Schmidtea mediterraneaFigure 1Comparison of the tRNA complement of Schistosoma mansoni Schistosoma japonicum and Schmidtea mediter-ranea A Comparison of anti-codon distributions for the 20 amino acids Numbers below each pie-chart are the total number of tRNA genes coding the corresponding amino acid Left columns S mansoni middle columns S mediterranea right columns S japonicum B Number of tRNAs encoding a particular amino acid red S mansoni blue S japonicum green S mediterranea Abbreviations Sup putative suppressor tRNAs (CTA TTA) Scys Selenocysteine tRNAs (TCA) Pseu predicted pseudo-genes Und tRNA predictions with uncertain anticodon likely these are also tRNA pseudogenes The Gln-tRNA derived repeat family (see text) is not included in these data C Comparison of codon usage and anti-codon abundance No significant correlation is observed for the two schistosomes For S mediterranea there is a weak but statistically significant positive cor-relation t asymp 20

TGCGGCCGCAGC

20 34 10 AlaTCCGCCCCCACC

31 27 5 GlyTGGGGGCGGAGG

48 50 12 Pro

TCTCCTTCGGCGCCGACG

58 44 13 Arg

GTGATG

27 8 2 His GCTACTTGAGGACGAAGA

51 94 19 Ser

GTTATT

23 27 3 AsnTATGATAAT

17 42 5 IleTGTGGTCGTAGT

35 34 7 Thr

GTCATC

8 15 5 Asp TAGGAGCAGAAGTAACAA

86 46 12 Leu

CCA23 23 0 Trp

GCAACA

21 44 5 CysTTTCTT

36 38 10 LysGTAATA

6 8 1 Tyr

TTGCTG

63 65 8 GlnCAT

21 44 7 MetTACGACCACAAC

37 29 13 Val

TTCCTC

39 44 8 GluGAAAAA

13 12 4 Phe

Sma Sme Sja Sma Sme Sja Sma Sme Sja

Ile Leu

S mansoniS mediterranea

S japonicum

000 002 004 006 008Fraction of Codons

SmansoniSjaponicumSmediterranea

acids are represented in approximately equal numbers inS mansoni and Schmidtea Nevertheless there are severalnotable deviations S mansoni contains many more leu-cine (86 vs 46) and histidine (27 vs 8) tRNAs while ser-ine (51 vs 94) cysteine (21 vs 44) methionine (21 vs44) and isoleucine (17 vs 42) are underrepresented Inaddition there are several substantial differences in codonusage In most cases S mansoni has a more diverse reper-toire of tRNAs tRNA-Asn-ATT tRNA-Arg-CGC tRNA-His-ATG tRNA-Ile-GAT tRNA-Pro-GGG tRNA-Tyr-ATAtRNA-Val-GAC are missing in Schmidtea Only tRNA-Ser-ACT is present in Schmidtea but absent in Schistosoma ThetRNA complement of S japonicum on the other hand dif-fers strongly from its two relatives Not only is the numberof tRNAs decreased by more than a factor of four S japon-icum also prefers anticodons that are absent or rare in itsrelatives such as tRNA-Ala-GGC tRNA-Cys-ACA and Lys-CTT On the other hand no tRNA-Trp was found Sincethe UGG codon is present in many open reading frameswe interpret this as a problem with the incompleteness ofthe genome assembly rather than a genuine gene loss Thereduction in the number of tRNAs is also evident by com-paring the number of tRNAs with introns 27 in S mansoniversus 5 in S japonicum

It has been shown recently that changes in codon usageeven while coding the same protein sequences canseverely attenuate the virulence of viral pathogens [36] byde-optimizing translational efficiency This observationleads us to speculate that the greater diversity of the tRNArepertoire could be related to the selection pressures of theparasitic life-style of S mansoni The effect is not straight-forward however because there is no significant correla-tion of tRNA copy numbers with the overall codon usagein both S mansoni and S japonicum Figure 1C In con-trast a weak but statistically significant correlation can beobserved in Schmidtea mediterranea It would be interest-ing therefore to investigate in detail whether there aredifferences in codon usage of proteins that are highlyexpressed in different stages of S mansonis life cycle andwhether the relative expression levels of tRNAs are understage-specific regulation

The most striking result of the tRNAscan-SE analysiswas the initial finding of 1135 glutamine tRNAs (Gln-tRNAs) in S mansoni in contrast to the 8 Gln-tRNAs in Sjaponicum and 65 in S mediterranea Nearly all of these(1098 in S mansoni) were tRNA-Gln-TTG In addition anextreme number of 1824 tRNA-pseudogenes in S man-soni (vs 951 in S japonicum and 19 in S mediterranea) waspredicted Of these 1270 were also homologous totRNA-Gln-TTG These two groups of tRNA-Gln-TTG-derived genes (those predicted to be pseudogenes andthose predicted to be functional tRNAs) totaled 2368These high numbers suggest a tRNA-derived mobile

genetic element We therefore ran the 2368 S mansonitRNA-Gln-TTG genes through the RepeatMasker pro-gram [37] Almost all of them (2342) were classified asSINE elements Further BLAST analysis revealed that theseelements are similar to members of the Sm-α family of Smansoni SINE elements [38] Removal of these SINE-likeelements yielded a total of 63 predicted glutamine-encod-ing tRNAs in S mansoni About 650 of 951 pseudogenesin S japonicum derived from tRNA-Pro-CGG

Homology-based analysis yielded similar though some-what less sensitive results to those of tRNAscan-SE Forinstance a BLAST search in S mansoni with Rfams tRNAconsensus yielded 617 predicted tRNAs compared to the663 predictions made by tRNAscan-SE

Ribosomal RNAsAs usual in eukaryotes the 18S 58S and 28S genes areproduced by RNA polymerase I from a tandemly repeatedpolycistronic transcript the ribosomal RNA operon TheS mansoni genome contains about 90-100 copies [3940]which are nearly identical at sequence level because theyare subject to concerted evolution [41] The repetitivestructure of the rRNA operons causes substantial prob-lems for genome assembly software [42] In order toobtain a conservative estimate of the copy number weretained only partial operon sequences that contained atleast two of the three adjacent rRNA genes We found 48loci containing parts of 18S 58S and 28S genes 32 locicovering 18S and 58S rRNA and 57 loci covering 58Sand 28S rRNAs [see Additional file 1 - Figures S1 and S2]Adding the copy numbers we have not fewer than 80 cop-ies (based on linked 18S rRNAs) and no more than 137copies (based on linked 58S rRNA) The latter is probablyan overestimate due to the possibility that the 58S rRNAmay be contained in two scaffolds The copy number ofrRNA operons is thus consistent with the estimate of 90-100 from hybridization analysis [39] An analogous anal-ysis of the current S japonicum assembly yields less accu-rate results Due to the many short fragments we obtained90 copies the true number may lie between 50 and 280however

The 5S rRNA is a polymerase III transcript that has notbeen studied in schistosomes so far We found 21 copiesof the 118 nt long 5S rRNA in S mansoni compared with13 copies in S japonicum Four of the 21 copies are locatedwithin a 3000 nt cluster on Scaffold010519

Spliceosomal RNAs and Spliced Leader RNASpliceosomes the molecular machines responsible formost splicing reactions in eukaryotic cells are ribonu-cleoprotein complexes similar to ribosomes [43] Themajor spliceosome which cleaves GT-AG intronsincludes the five snRNAs U1 U2 U4 U5 and U6 In the

S mansoni genome all of them are multicopy genes Byhomology search we found 34 U1 15 U2 19 U4 9 U5and 55 U6 sequences in the genome assembly Interpret-ing all sequences that are identical in short flankingregions as the same we would retain only 3 U1 3 U2 1U4 2 U5 and nine U6 genes [44] The true copy numberin the S mansoni genome is most likely somewherebetween these upper and lower bounds For S japonicumthe corresponding numbers are U1 2-6 U2 1-63 U2 U41-6 U4 U5 1-24 and U6 2-12 Due to the more frag-mented genome assembly we expect the true numbers tobe closer to the lower bounds Secondary structures forthese candidates are similar to those of typical snRNAsFigure 2

A second much less frequent minor spliceosome isresponsible for the processing of atypical AT-AC intronsIt shares only the U5 snRNA with the major spliceosomeThe other four RNA components are replaced by variantscalled U11 U12 U4atac and U6atac [45] The minor-spliceosomal snRNAs are typically much less conservedthan the RNA components of the major spliceosome [44]It was not surprising therefore that these RNAs weredetectable only by means of GotohScan[8] but not withthe much less sensitive BLAST searches Although U4atacand U6atac are quite diverged compared to known

homologs they can be recognized unambiguously basedon both secondary structure and conserved sequencemotifs Furthermore the U4atac and U6atac sequencescan interact to form the functionally necessary duplexstructure shown in Figure 2 As in many other speciesthere is only a single copy of each of the minor spliceo-somal snRNAs in both of the schistosome genomes Tab1 An analysis of promoter sequences showed that theputative snRNA promoter motifs in S mansoni are highlyderived Only one of the two U12 genes exhibited a clearlyvisible snRNA-like promoter organization

The Spliced Leader (SL) RNA is one of the very few previ-ously characterized ncRNAs from S mansoni [26] The 90nt SL RNA which was found in a 595 nt tandemlyrepeated fragment (accession number M34074) containsthe 36 nt leader sequence at its 5 end which is transferredin the trans-splicing reaction to the 5 termini of maturemRNAs Using blastn we identified 54 SL RNA genesThese candidates along with 100 nt flanking sequencewere aligned using ClustalX revealing 6 sequences withaberrant flanking regions which we suspect to be pseu-dogenic The remaining sequences are 43 identical copiesand 5 distinct sequence variants A secondary structureanalysis corroborates the model of [26] according towhich the S mansoni SL RNA has only two loops with an

Secondary structures of the nine snRNAs and the interaction complexes of U4U6 and U4atacU6atac respectively in S man-soniFigure 2Secondary structures of the nine snRNAs and the interaction complexes of U4U6 and U4atacU6atac respec-tively in S mansoni Structure prediction was performed by RNAfold RNAalifold and for U4U6 and U4atacU6atac by RNAcofold from the RNA Vienna Package [96108] Boxes indicate Sm binding sites Additional details on sequences structures and alignments are available at the supplementary material

5rsquo

3rsquo

5rsquo 3rsquo

3rsquo5rsquo

5rsquo3rsquo

5rsquo 3rsquo

5rsquo

5rsquo3rsquo

3rsquo

5rsquo

5rsquo3rsquo

U4atacU6atac

unpaired Sm binding site [see Additional File 1 - FigureS3] This coincides with the SL RNA structure of Rotifera[46] but is in contrast to the SL RNAs in most othergroups of eukaryotes which exhibit single or triple stem-loop structures [47] A blastn-search against S mansoniEST data confirms that the 5 part of the SL is indeed trans-spliced to mRNAs Several nearly identical SL RNAhomologs are found in S japonicum

SRP RNA and Ribonuclease P RNASignal recognition particle (SRP) RNA also known as 7SLRNA is part of the signal recognition particle a ribonucle-oprotein that directs packaged proteins to their appropri-ate locations in the endoplasmic reticulum Although oneof the protein subunits of this ribonucleoprotein wascloned in 1995 [48] little is known about the other subu-nits or the RNA component in S mansoni We found eightprobable candidates for the SRP RNA with one almostcanonical sequence [see Additional file 1 - Figure S4] andfour possible candidates with point mutations which mayinfluence their function

The RNA component of Ribonuclease P (RNase P) is thecatalytically active part of this enzyme that is required forthe processing of tRNA precursors [4950] We found oneclassic RNase P RNA in the S mansoni genome using bothGotohScan and rnabob with the eukaryotic (nuclear)Rfam consensus sequence for RNase P as search sequence

MicroRNAsMicroRNAs are small RNAs that are processed from hair-pin-like precursors see eg [51] They are involved inpost-transcriptional regulation of mRNA molecules Sofar no microRNAs have been verified experimentally in Smansoni The presence of four protein-coding genesencoding crucial components of the microRNA process-ing machinery (Dicer Argonaut Drosha and PashaDGCR8) [5253] and the presence of Argonaut-like genesin both S japonicum [54] and S mansoni (detected bytblastn in EST data see Supplemental Data online)strongly suggests that schistosomes have a functionalmicroRNA system Indeed most recently five miRNAswere found by direct cloning in S japonicum that are alsoconserved in S mansoni [55] let-7 mir-71 bantam mir-125 and a single schistosome-specific microRNA Thesesequences including the precursor hairpins are well con-served in S japonicum On the other hand the microRNAprecursor sequences of both schistosomes are quitediverged from the consensus of the homologous genes inBilateria

Using bioinformatics (see methods) we were able to findonly one further miRNA candidate in S mansoni mir-124that is also conserved in S japonicum In insects thismiRNA is clustered with mir-287 The distance of both

miRNAs is approximately 8 kb in Drosophilids We foundan uncertain mir-287 candidate in S mansoni however ona different scaffold than mir-124 Although this sequencenicely folds into a single stem-loop structure it is con-served only antisense to the annotated mature sequencein insects (see Figure 3) This S mansoni mir-287 candi-date does not seem to be conserved in S japonicum

In [56] 71 microRNAs are described for the distantlyrelated trematode Schmidtea mediterranea and additionalones are announced in a recent study focussing on piRNAs[57] The overwhelming majority 54 were reported to bemembers of 29 widely conserved metazoan microRNAfamilies although in some cases even the mature miRNAsequence is quite diverged Therefore we regard severalfamily assignments as tentative at best Of those 29 miR-NAs we found mir-124 only However the schistosomesequences are more related to the other bilaterian mir-124homologs than to those of S mediterranea Out of theremaining 54 miRNAs that were annotated in S mediter-ranea we found that mir-749 is also conserved in the twoschistosome species Here the sequences show a commonconsensus sequence and secondary structure in their pre-cursors (see Figure 3)

The small number of recognizable microRNAs in schisto-somes is in strong contrast to the extensive microRNAcomplement in S mediterranea indicating massive loss ofmicroRNAs relative to the planarian ancestor This may bea consequence of the parasitic lifestyle of the schisto-somes

Small Nucleolar RNAsSmall nucleolar RNAs play essential roles in the process-ing and modification of rRNAs in the nucleolus [5859]Both major classes the box HACA and the box CD snoR-NAs are relatively poorly conserved at the sequence leveland hence are difficult to detect in genomic sequencesThis has also been observed in a recent ncRNA annotationproject of the Trichoplax adhaerens genome [8] The best-conserved snoRNA is the atypical U3 snoRNA which isessential for processing of the 18S rRNA transcript intomature 18S rRNA [60] In the current assembly of the Smansoni genome we found six U3 loci but they are alsoidentical in the flanking sequences suggesting that in factthere is only a single U3 gene No unambiguous homo-logue was detected for any of the other known snoRNAs

A de novo search for snoRNAs (see methods for details)resulted in 2610 promising candidates (1654 box CDand 956 box HACA) see Supplemental Data online Allthese predictions exhibit highly conserved sequence boxesas well as the typical secondary features of box CD andbox HACA snoRNAs respectively

A comparison of the predicted snoRNAs with the entriesin the Rfam[61] and NONCODE[62] databases returnedonly 47 hits that match to several other RNAs like tRNAsparts of the rRNA operon snRNAs mRNAlike genes and afew of our candidates map to the hammerhead ribozymeThese sequences are likely false positives and have beenremoved from the candidate list The number of predictedcandidates is much larger than the number of snoRNAsreported in other organisms for instance [59] lists 456 forthe human genome Although we most likely do not yetknow the full snoRNA complement of eukaryoticgenomes we have to expect that a large fraction of predic-tion will turn out to be false positives

We therefore analysed the conservation of the candidatesin S japonicum and focussed on the snoRNA candidateswith targets in the 18S 28S andor 58S ribosomal RNAWhile targets are predicted for more than half of the can-didates see Table 2 the numbers are drastically reducedwhen conservation of the candidates in S japonicum isrequired Note furthermore that the fraction of con-served candidates is strongly enriched among those withribosomal RNA targets indicating that these sets are likelyto contain a sizeable fraction of true positives This filter-ing step leaves us with 227 box CD and 352 box HACAsnoRNA candidates While still high these numbers fallinto the expected range for a metazaon snoRNA comple-ment

Multiple sequence alignments of the pre-miRNAs that were computationally found in S mansoniFigure 3Multiple sequence alignments of the pre-miRNAs that were computationally found in S mansoni For mir-124 and mir-749 the sequences share a common consensus structure The uncertain mir-287 candidate clusters together with mir-124 in insect genomes However though it also exhibits a single stem-loop structure it is different from that of insects Here the sequence is only conserved at the antisense region of the annotated mature miRNA

Structure (((((((((((((((((((((((())))))))))))))))))))))))sma-mir-124 UUGUAUGCCAUUUUCCGCGAUUGCCUUGAUGAGUUAUAA--AUAUUAUUCAUAACAAAAAUAUUAAGGCACGCGGUGAAUGUCAUCCACGGsja-mir-124 AUGUAUGCCAUUUUCCGCGAUUGCCUUGAUUUGUUAAAAGAAAAUGAUUCACAACAAAA-UAUUAAGGCACGCGGUGAAUGUCAUCCACGGhsa-miR-124 ---------------------------------------------------------------UAAGGCACGCGGUGAAUGCC--------

mir-124

|-conserved antisense--| dme-Struc (((((())))))(((((((((((((((((((((((((())))))))))))))))))))))))))dme-mir-287 GGACGCCGGGGAUGUAUGGG--UGUGUA--GGGUCUGAAAUUUUGCACACAUUUACAAUAAUUGUAAAUGUGUUGAAAAUCGUUUGCACGACUGUGAdme-miR-287 --------------------------------------------------------------------UGUGUUGAAAAUCGUUUGCAC--------sma-mir-287 ---GUAUACUCGUAUGGGUGAAUGUGUACA---UGUUAAAUUUUGCACACAUUUACAAAAAAAAGGUGCCGAAUAUUCCAUUUUCACCCUACAUGUUsma-Struc (((((((((((((((((((((((((())))))))))))))))))))))))))

mir-287

sme-miR-749 Structure (((((((((((((((((((((((((((((())))))))))))))))))))))))))))))sja-mir-749 AAUCGCCAGGAUGAACCUCGGUGGUCCGGGGUGCAGGCUUCAAACCUGCAGCCGACUGGCGUCGGAGUGGUUCGAUUCCGCCUUCCUGGCGUGsma-mir-749 AAUUGCCGGGAUGAACCUCGGUGGUCCGGGGUGCAGGCUUCAAACCUGUAGCCGACUAGCAUCGGAGCGGUUCGAUUCCGCCUUCCUGGCGUAsme-mir-749-1 AAUCGCUGGGAUGAGCCUCGGUGGUCCGGGGUGCAGGCUUCAAACCUGUAGUCGGUUGACACCGAAGUGGUUCGAUUCCACCUUUCCAGCGAUsme-mir-749-2 AAUUGCUGGGAUGAGCCUCGGUGGUCCGGGGUGCAGGCUUCAAACCUGUAGUCGGUUGACACCGAAGUGGUUCGAUUCCACCUUUCCAGCGAUsme-miR-749 ----GCUGGGAUGAGCCUCGGUGGU--------------------------------------------------------------------

mir-749

Table 2 Conservation and target prediction of snoRNA candidates

snoReporttargets Box CD (snoscan) Box HACA (RNAsnoop)ge 2 1 0 ge 2 1 0

predicted in S mansoni 926 110 613 284 495 177conserved in S japonicum 200 27 83 149 203 62

Only ribosomal RNAs were searched for putative target sites

We remark finally that five of the snoRNA candidates(three box CD and two box HACA) are also conserved inSchmidtea mediterranea

Other RNA motifsTwo examples of relatively well-known schistosome non-coding RNAs are the hammerhead ribozyme motifswithin the Sm-α and Sj-α SINE-like elements [2728] Ablastn search of the hammerhead ribozyme motif fromthe Rfam database resulted in ~38500 candidates for Smansoni in contrast to ~5000 candidates for S japonicumWhile high this number is not surprising considering thegenerally high copy number of SINE elements previouslythe copy number for Sm-α elements in the S mansonigenome was estimated to exceed 10000 [27] The highlyconserved potassium channel RNA editing signal [6364]is another structured RNA element that was described pre-viously [65] We found nine copies of this hairpin struc-ture in the S mansoni genome assembly and three in Sjaponicum

Uncertain and missing candidatesBoth the MRP RNA [2366] and the 7SK RNA [4567]have highly variable rapidly evolving sequences thatmake them difficult or impossible to detect in invertebrategenomes Their ancient evolutionary origin and theirextremely conserved molecular house-keeping functionsmake it more than likely that they are present in the schis-tosome genomes as well In both cases we have not beenable to identify unambiguous homologs There are how-ever plausible candidates We briefly describe them in thefollowing paragraphs since they may warrant furtherattention and may be a useful starting point for subse-quent experimental studies as exemplified by the historyof discovery of the snRNA in Giardia intestinalis [68-70]

MRP RNA has multiple functions among them mito-chondrial RNA processing and nucleolar pre-rRNAprocessing The S mansoni MRP candidate fits the generalsecondary structure model of metazoan MRP RNAs[2366] and analysis with RNAduplex shows that thecandidate contains a pseudoknot which exhibited strikingsequence identity with known MRPs The locus is well-conserved in S japonicum On the other hand stems 1 and12 were divergent compared to known MRPs and stem 19also fails to display clear similarities with known MRPsAlthough quite likely a true MRP homolog we thereforeconsider this sequence only tentative

7SK RNA is a general transcriptional regulator repressingtranscript elongation through inhibition of transcriptionelongation factor PTEFb and also suppresses the deami-nase activity of APOBEC3C [71] The S mansoni 7SK can-didate has a 5 stem similar to that described in otherinvertebrates [5] and parts of the middle of the sequenceare also recognizable There is furthermore a homolo-

gous locus in the genome of S japonicum However the 3stem (which was followed by a poly-T terminator) was notconserved In addition a large sequence deletion was evi-dent

Three major classes of ncRNAs were expected but notfound in the S mansoni genome As in all other inverte-brates genomes no candidate sequence was found for atelomerase RNA S mansoni almost certainly has a canon-ical telomerase holoenzyme since it encodes telomeraseproteins (Smp_066300 and Smp_066290) and has thesame telomeric repeat sequences as many other metazoananimals [72] Telomerase RNAs are notoriously difficultto find as they are highly divergent among different spe-cies varying in both size and sequence composition[773] Vault RNAs are known in all major deuterostomelineages [74] and homologs were recently also describedin two lophotrochozoan lineages [75] Since S mansonihas a homolog of the major vault protein (Smp_006740)we would also expect a corresponding RNA component tobe present So far Y RNAs have been found only in verte-brates [7677] and in nematodes [7879] although the RoRNP that they are associated with seems to be present inmost or even all eukaryotes

ConclusionWe have described here a detailed annotation of house-keeping ncRNAs in the genomes of the parasitic platy-helminth Schistosoma mansoni and Schistosoma japonicumLimited to the best conserved structured RNAs our worknevertheless uncovered important genomic features suchas the existence of a SINE family specific to Schistosomamansoni which is derived from tRNA-Gln-TTG Our datafurthermore establish the presence of a minor spliceo-some in schistosomes and confirms spliced leader trans-splicing With a coverage of at least 90-95 of thegenomic DNA missing data are not a significant problemThe fragmented genome assemblies however precludeaccurate counts of the multi-copy genes

Platyhelminths are known to be a fast-evolving phylum[80] It is not surprising therefore that in particular thesmall ncRNAs are hard or impossible to detect by simplehomology search tools such as blastn Even specializedtools have been successful in identifying only the betterconserved genes such as tRNA microRNAs RNase P RNASRP RNA The notoriously poorly conserved familiessuch as snoRNAs telomerase RNA or vault RNAs mostlyescaped detection

The description of several novel and in many cases quitederived schistosome ncRNAs contributes significantly tothe understanding of the evolution of the correspondingRNA families The schistosome ncRNA sequences further-more are an important input to subsequent homologysearch projects since they allow the construction of

improved descriptors for sequencestructure-based searchalgorithms Last but not least the ncRNA annotationtracks are an important contribution to the genome-wideannotation datasets of both S mansoni and S japonicumIt not only contributes the protein-based annotation butalso helps to identify annotation errors eg cases whereputative proteins are annotated that overlap rRNA oper-ons or other ncRNAs

The house-keeping ncRNAs considered in this study arealmost certainly only the proverbial tip of the platy-helminth ncRNAs iceberg The discovery of a largenumber of mRNA-like ncRNAs (mlncRNAs) in manyeukaryotes (compiled eg in the RNAdb[81] and reviewedeg in [1]) and in particular in many other invertebratespecies (nematodes [82] insects [8384]) suggests thatsimilar transcripts will also be abundant in schistosomesThe abundant EST data for both schistosome species[8586] can provide a starting point eg for an analysisalong the lines of [87] Computational surveys further-more have provided evidence for large numbers of RNAswith conserved secondary structures in other invertebrates[88-90] The underlying methods such as RNAz[91] areinherently comparative presenting difficulties for appli-cation to schistosome genomes due to the large evolution-ary distance between schistosome and non-schistosomegenomes This is also the case for a recent approach toidentify mRNA-like non-coding RNAs with very low levelsof sequence conservation based on their intron structure[92] A deeper understanding of the non-coding transcrip-tome of schistosomes will therefore have to rely primarilyon experimental approaches either by means of tilingarrays or by means of high throughput transcriptomesequencing

MethodstRNA annotationWe used tRNAscan-SE[93] with default parameters toannotate putative tRNA genes As additional confirma-tion the genome sequence was searched using tRNA con-sensus sequences from the Rfam database [61] In order toobtain suitable data for comparison the genome of thefree-living platyhelminth Schmidtea mediterranea [17] wassearched alongside that of S mansoni and S japonicum

microRNA annotationWe followed the general protocol outlined in [8] to iden-tify miRNA precursors using all metazoan miRNAs listedin miRBase [94] [Release 110 httpmicrornasangeracuksequences] The initial search was con-ducted by blastn with E lt 001 with the mature andmature miRNAs as query sequences The resulting candi-dates were then extended to the length of the precursorsequence of the search query and aligned to the precursorsusing ClustalW[95] Secondary structures were pre-dicted using RNAfold[96] for single sequences and

RNAalifold[97] for alignments Candidates that didnot fold into miRNA-like hairpin structures were dis-carded The remaining sequences were then examined byeye to see if the mature miRNA was well-positioned in thestem portion of each putative precursor sequence In addi-tion we used the final candidates to search the S japoni-cum and S mediterranea genomes to examine whetherthese sequences are conserved in Schistosoma andorPlatyhelminthes

snoRNA annotationWe compared all the known human and yeast snoRNAsthat are annotated in the snoRNAbase[98] to the S man-soni genome using BLAST[99] and GotohScan[8] Thesearch for novel snoRNA candidates was performed onlyon sequences that were not annotated as protein-codingor another ncRNA in the current S mansoni assembly TheSnoReport program [100] was used to identify putativebox CD and box HACA snoRNAs on both strands Onlythe best predictions ie those that show highly conservedboxes and canonical structural motifs were kept for fur-ther analysis The remaining candidates are further ana-lysed for possible target interactions with ribosomal RNAsusing snoscan[101] for box CD and RNAsnoop[102]for box HACA snoRNA candidates In addition thesequences were checked for conservation in S japonicumand S mediterranea using BLAST To estimate the numberof false predictions we compared the candidate snoRNAswith common ncRNA databases in particular Rfam[61]and NONCODE[62] All sequences that match a non-snoRNA ncRNA were discarded

Other RNA familiesFor other families we employed the following five steps

(a) For candidate sequences of ribosomal RNAs spliceo-somal RNAs the spliced leader (SL) and the SRP RNA weperformed BLAST searches with E lt 0001 using theknown ncRNA genes from the NCBI and Rfam databasesFor the snRNA set see [44] For 7SL RNA we used X04249for 5S and 58S rRNAs we used the complete set of Rfamentries for the SSU and LSU rRNAs we used Z11976 andNR_003287 respectively The SL RNAs were searchedusing SL RNA entries from Rfam and the sequencesreported in [26] For more diverged genes such as minorsnRNAs RNase MRP 7SK and RNase P we usedGotohScan[8] an implementation of a full dynamicprogramming alignment with affine gap costs In caseswhere no good candidates were found we also employeddescriptor-based search tools such as rnabob httpselabjaneliaorgsoftwarehtml

(b) In a second step known and predicted sequences werealigned using ClustalW[95] and visualized with ClustalX[103] To identify functional secondary structureRNAfold RNAalifold and RNAcofold[104] were

used Combined primary and secondary structures werevisualized using stockholm-format alignment files inthe emacs editor utilizing ralee mode [105] Align-ments are provided at the Supplemental Data online

(c) Putatively functional sequences were distinguishedfrom likely pseudogenes by analysis of flanking genomicsequence To this end the flanking sequences of snRNAand SL RNA copies were extracted and analyzed for con-served sequence elements using MEME[106] Only snRNAswith plausible promoter regions were reported

(d) Additional consistency checks were employed forindividual RNA families including phylogenetic analysisby neighbor-joining [107] to check that candidatesequences fall at phylogenetically reasonable positionsrelative to previously known homologs For RNase MRPRNA candidates RNAduplex httpwwwtbi univieacatRNARNAduplexhtml was used to find the pseu-doknot structure In order to confirm that the SL RNA can-didate is indeed trans-spliced to mRNA transcripts wesearched the FAPESP Genoma Schistosoma mansoni websitefor ESTs including fragments of the predicted SL RNA Wefound 52 ESTs with blastnE lt 0001 that span the pre-dicted region of the SL RNA (nt 8-38) indicating that thisRNA does indeed function as a spliced leader

(e) Accepted candidate sequences were used as BLASTqueries against the S mansoni genome to determine theircopy number in the genome assembly

Additional Data OnlineThe website httpwwwbioinfuni-leipzigdePublicationsSUPPLEMENTS08-014 provides extensive machinereadable information including sequence files align-ments and genomic coordinates

Authors contributionsCSC PB and PFS designed the study CSC MM DR JHCBS SK CSA and PFS performed the computationalanalyses CSC wrote the first draft of the manuscript Allauthors contributed to the final assessment of the data aswell as the writing of the final version of the manuscriptCSC MM DR JH should be considered as joint firstauthors

Additional material

AcknowledgementsThis work was supported in part by the European Union through grants in the 6th and 7th Framework Programe of the European Union (projects EMBIO SYNLET and EDEN) the Deutsche Forschungsgemeinschaft and the auspices of SPP SPP-1174 Deep Metazoan Phylogeny the Freistaat Sach-sen and the DAAD-AleCol program

References1 Amaral PP Dinger ME Mercer TR Mattick JS The eukaryotic

genome as an RNA machine Science 2008 3191787-17892 Piccinelli P Rosenblad MA Samuelsson T Identification and anal-

ysis of ribonuclease P and MRP RNA in a broad range ofeukaryotes Nucleic Acids Res 2005 334485-4495

3 Woodhams MD Stadler PF Penny D Collins LJ RNAse MRP andthe RNA Processing Cascade in the Eu-karyotic AncestorBMC Evol Biol 2007 7S13

4 Gruber AR Koper-Emde D Marz M Tafer H Bernhart S Obernos-terer G Mosig A Hofacker IL Stadler PF Benecke BJ Invertebrate7SK snRNAs J Mol Evol 2008 107-11566

5 Gruber A Kilgus C Mosig A Hofacker IL Hennig W Stadler PFArthropod 7SK RNA Mol Biol Evol 2008 1923-193025

6 Chen JL Blasco MA Greider CW Secondary Structure of Ver-tebrate telomerase RNA Cell 2000 100503-514

7 Xie M Mosig A Qi X Li Y Stadler PF Chen JJL Size Variation andStructural Conservation of Vertebrate Telomerase RNA JBiol Chem 2008 2832049-2059

8 Hertel J de Jong D Marz M Rose D Tafer H Tanzer A SchierwaterB Stadler PF Non-Coding RNA Annotation of the Genome ofTrichoplax adhaerens Nucleic Acids Res 2009 371602-1615

9 Blair D Davis GM Wu B Evolutionary relationships betweentrematodes and snails emphasizing schistosomes and parag-onimids Parasitology 2001 123(Suppl)S229-S243

10 Brant SV Loker ES Can specialized pathogens colonize dis-tantly related hosts Schistosome evolution as a case studyPLoS Pathog 2005 1167-169

11 Webster BL Southgate VR Littlewood DTJ A revision of theinterrelationships of Schistosoma including the recentlydescribed Schistosoma guineensis Int J Parasitol 200636947-955

12 Jimeacutenez-Guri E Philippe H Okamura B Holland PWH Buddenbroc-kia is a cnidarian worm Science 2007 317116-118

13 Wilson RA Ashton PD Braschi S Dillon GP Berriman M Ivens AOming in on schistosomes prospects and limitations forpost-genomics Trends Parasitol 2007 2314-20

14 Berriman M Haas BJ LoVerde PT Wilson RA Dillon GP CerqueiraGC Mashiyama ST Al-Lazikani B Andrade LF Ashton PD Aslett MABartholomeu DC Blandin G Caffrey CR Coghlan A Coulson R DayTA Delcher A DeMarco R Djikeng A Eyre T Gamble JA Ghedin EGu Y Hertz-Fowler C Hirai H Hirai Y Houston R Ivens A JohnstonDA Lacerda D Macedo CD McVeigh P Ning Z Oliveira G Overing-ton JP Parkhill J Pertea M Pierce RJ Protasio AV Quail M Rajan-dream MA Rogers J Sajid M Salzberg SL Stanke M Tivey AR WhiteO Williams DL Wortman J Wu W Zamanian M Zerlotini A Fraser-Liggett CM Barrell BG El-Sayed NM The genome of the bloodfluke Schistosoma mansoni Nature 2009 460352-358

15 Schistosoma japonicum Genome Sequencing and Functional AnalysisConsortium The Schistosoma japonicum genome reveals fea-tures of host-parasite interplay Nature 2009 460345-351

16 Hirai H Taguchi T Saitoh Y Kawanaka M Sugiyama H Habe SOkamoto M Hirata M Shimada M Tiu WU Lai K Upatham ES Agat-suma T Chromosomal differentiation of the Schistosomajaponicum complex Int J Parasitol 2000 30441-452

17 Robb SMC Ross E Alvarado AS SmedGD the Schmidtea mediter-ranea genome database Nucleic Acids Res 2008 36D599-D606

18 Haas BJ Berriman M Hirai H Cerqueira GG Loverde PT El-SayedNM Schistosoma mansoni genome closing in on a final geneset Exp Parasitol 2007 117225-228

19 Hu W Yan Q Shen DK Liu F Zhu ZD Song HD Xu XR Wang ZJRong YP Zeng LC Wu J Zhang X Wang JJ Xu XN Wang SY Fu GZhang XL Wang ZQ Brindley PJ McManus DP Xue CL Feng ZChen Z Han ZG Evolutionary and biomedical implications ofa Schistosoma japonicum complementary DNA resource NatGenet 2003 35139-147

Additional file 1Supplemental figures and captions contains supplemental Figures S1 - S4 mentioned in the main textClick here for file[httpwwwbiomedcentralcomcontentsupplementary1471-2164-10-464-S1PDF]

20 Verjovski-Almeida S R D Martins EA Guimaratildees PE Ojopi EPPaquola AC Piazza JP Nishiyama MY Jr Kitajima JP Adamson REAshton PD Bonaldo MF Coulson PS Dillon GP Farias LP GregorioSP Ho PL Leite RA Malaquias LC Marques RC Miyasato PA Nasci-mento AL Ohlweiler FP Reis EM Ribeiro MA Saacute RG Stukart GCSoares MB Gargioni C Kawano T Rodrigues V Madeira AM WilsonRA Menck CF Setubal JC Leite LC Dias-Neto E Transcriptomeanalysis of the acoelomate human parasite Schistosoma man-soni Nat Genet 2003 35148-157

21 Verjovski-Almeida S Venancio TM Oliveira KC Almeida GTDeMarco R Use of a 44k oligoarray to explore the transcrip-tome of Schistosoma mansoni adult worms Exp Parasitol 2007117236-245

22 Schulmeister A Heyers O Morales ME Brindley PJ Lucius R MeuselG Kalinna BH Organization and functional analysis of theSchistosoma mansoni cathepsin D-like aspartic protease genepromoter Biochim Biophys Acta 2005 172727-34

23 Copeland CS Mann VH Brindley PJ Both sense and antisensestrands of the LTR of the Schistosoma mansoni Pao-like ret-rotransposon Sinbad drive luciferase expression Mol GenetGenomics 2007 277161-170

24 Brejovaacute B Vinaz T Chen Y Wang S Zhoa G Brown DG Li M ZhouY Finding genes in Schistosoma japonicum annotating novelgenomes with help of extrinsic evidence Nucleic Acids Res 200937e52

25 Mourier T Carret C Kyes S Christodoulou Z Gardner PP JeffaresDC Pinches R Barrell B Berriman M Griffiths-Jones S Ivens A New-bold C Pain A Genome-wide discovery and verification ofnovel structured RNAs in Plasmodium falciparum Genome Res2008 18281-292

26 Rajkovic A Davis RE Simonsen JN Rottman FM A spliced leaderis present on a subset of mRNAs from the human parasiteSchistosoma mansoni Proc Natl Acad Sci USA 1990 878879-8883

27 Ferbeyre G Smith JM Cedergren R Schistosome satellite DNAencodes active hammerhead ribozymes Mol Cell Biol 1998183880-3888

28 Laha T McManus DP Loukas A Brindley PJ Sjα elements shortinterspersed element-like retroposons bearing a hammer-head ribozyme motif from the genome of the oriental bloodfluke Schistosoma japonicum Biochim Biophys Acta 20001492477-482

29 Copeland CS Heyers O Kalinna BH Bachmair A Stadler PFHofacker IL Brindley PJ Structural and evolutionary analysis ofthe transcribed sequence of Boudicca a Schistosoma mansoniretrotransposon Gene 2004 329103-114

30 Rollinson D Kaukas A Johnston DA Simpson AJ Tanaka M Somemolecular insights into schistosome evolution Int J Parasitol1997 2711-28

31 Littlewood DT Lockyer AE Webster BL Johnston DA Le TH Thecomplete mitochondrial genomes of Schistosoma haemato-bium and Schistosoma spindale and the evolutionary history ofmitochondrial genome changes among parasitic flatwormsMol Phylogenet Evol 2006 39452-467

32 DeMarco R Verjovski-Almeida S Expressed Sequence Tags(ESTs) and Gene Discovery Schistosoma mansoni Bioinformat-ics in Tropical Disease Research A Practical and Case-Study Approach2008B06 [httpwwwncbinlmnihgovbookshelfbrfcgibook=bioinfo] Bethesda MD National Library of Medicine

33 Sheppard K Akochy PM Soumlll D Assays for transfer RNA-dependent amino acid biosynthesis Methods 2008 44139-145

34 Ambrogelly A Palioura S Soumlll D Natural expansion of thegenetic code Nat Chem Biol 2007 329-35

35 Hubert N Walczak R Sturchler C Myslinski E Schuster C WesthofE Carbon P Krol A RNAs mediating cotrans-lational insertionof selenocysteine in eukaryotic selenoproteins Biochimie 199678590-596

36 Coleman JR Papamichail D Skiena S Futcher B Wimmer E MuellerS Virus attenuation by genome-scale changes in codon pairbias Science 2008 3201784-1787

37 Smit AFA Hubley R Green P RepeatMasker [Version open-325 [RMLib 20080611] [httpwwwrepeatmaskerorg]

38 Spotila LD Hirai H Rekosh DM Lo Verde PT A retroposon-likeshort repetitive DNA element in the genome of the humanblood fluke Schistosoma mansoni Chromosoma 198997421-428

39 Simpson AJ Dame JB Lewis FA McCutchan TF The arrangementof ribosomal RNA genes in Schistosoma mansoni Identifica-tion of polymorphic structural variants Eur J Biochem 198413941-45

40 van Keulen H Loverde PT Bobek LA Rekosh DM Organization ofthe ribosomal RNA genes in Schistosoma mansoni Mol Bio-chem Parasitol 1985 15215-230

41 Nei M Rooney AP Concerted and birth-and-death evolutionof multigene families Annu Rev Genet 2005 39121-152

42 Scheibye-Alsing K Hoffmann S Frankel AM Jensen P Stadler PFMang Y Tommerup N Gilchrist MJ Hillig ABN Cirera S JoslashrgensenCB Fredholm M Gorodkin J Sequence Assembly Comp BiolChem 2009 33121-136

43 Staley JP Woolford JL Jr Assembly of ribosomes and spliceo-somes complex ribonucleoprotein machines Curr Opin CellBiol 2009 21109-118

44 Marz M Kirsten T Stadler PF Evolution of Spliceosomal snRNAGenes in Metazoan Animals J Mol Evol 2008 67594-607

45 Kreivi JP Lamond AI RNA splicing unexpected spliceosomediversity Curr Biol 1996 6802-805

46 Pouchkina-Stantcheva NN Tunnacliffe A Spliced leader RNA-mediated trans-splicing in phylum Rotifera Mol Biol Evol 2005221482-1489

47 Marz M Vanzo N Stadler PF Carnival of SL RNAs Structuralvariants and the possibility of a common origin J Bioinf CompBiol 2009 [httpwwwbioinfuni-leipzigdePublicationsPREPRINTS09-009pdf]

48 McNair A Zemzoumi K Luumltcke H Guillerm C Boitelle A Capron ADissous C Cloning of a signal-recognition-particle subunit ofSchistosoma mansoni Parasitol Res 1995 81175-177

49 Kirsebom LA RNase P RNA mediated cleavage substrate rec-ognition and catalysis Biochimie 2007 891183-1194

50 Kikovska E Svaumlrd SG Kirsebom LA Eukaryotic RNase P RNAmediates cleavage in the absence of protein Proc Natl Acad SciUSA 2007 1042062-2067

51 Williams AE Functional aspects of animal microRNAs Cell MolLife Sci 2008 65545-562

52 Krautz-Peterson G Skelly PJ Schistosoma mansoni the dicergene and its expression Exp Parasitol 2008 118122-128

53 Gomes MS Cabral FJ Jannotti-Passos LK Carvalho O Rodrigues VBaba EH Saacute RG Preliminary analysis of miRNA pathway inSchistosoma mansoni Parasitol Int 2009 5861-68

54 Liu F Lu J Hu W Wang SY Cui SJ Chi M Yan Q Wang XR SongHD Xu XN Wang JJ Zhang XL Zhang X Wang ZQ Xue CL Brind-ley PJ McManus DP Yang PY Feng Z Chen Z Han ZG New per-spectives on host-parasite interplay by comparativetranscriptomic and proteomic analyses of Schistosoma japon-icum PLoS Pathog 2006 2e29

55 Xue X Sun J Zhang Q Wang Z Huang Y Pan W Identificationand characterization of novel microRNAs from Schistosomajaponicum PLoS ONE 2008 3e4034

56 Palakodeti D Smielewska M Graveley BR MicroRNAs from thePlanarian Schmidtea mediterranea a model system for stemcell biology RNA 2006 121640-1649

57 Palakodeti D Smielewska M Lu YC Yeo GW Graveley BR ThePIWI proteins SMEDWI-2 and SMEDWI-3 are required forstem cell function and piRNA expression in planarians RNA2008 141174-1186

58 Matera AG Terns R Terns Non-coding RNAs lessons from thesmall nuclear and small nucleolar RNAs Nat Rev Mol Cell Biol2007 8209-220

59 Dieci G Preti M Montanini B Eukaryotic snoRNAs A paradigmfor gene expression flexibility Genomics 2009 9483-88

60 Lukowiak AA Granneman S Mattox SA Speckmann WA Jones KPluk WJ Venrooij Hand Terns RM Terns MP Interaction of theU3-55k protein with U3 snoRNA is mediated by the box BCmotif of U3 and the WD repeats of U3-55k Nucleic Acids Res2000 283462-3471

61 Griffiths-Jones S Moxon S Marshall M Khanna A Eddy SR BatemanA Rfam annotating non-coding RNAs in complete genomesNucleic Acids Res 2005 33D121-D124

62 Liu C Bai B Skogerboslash G Cai L Deng W Zhang Y Bu DB Zhao YChen R NONCODE an integrated knowledge database of non-coding RNAs Nucleic Acids Res 2005 33D112-D115

63 Bhalla T Rosenthal JJC Holmgren M Reenan R Control of humanpotassium channel inactivation by editing of a small mRNAhairpin Nature Struct Mol Biol 2004 11950-956

64 Yang Y Lv J Gui B Yin H Wu X Zhang Y Jin Y A-to-I RNA edit-ing alters less-conserved residues of highly conserved codingregions implications for dual functions in evolution RNA2008 141516-1525

65 Kim E Day TA Bennett JL Pax RA Cloning and functionalexpression of a Shaker-related voltage-gated potassiumchannel gene from Schistosoma mansoni (Trematoda Dige-nea) Parasitology 1995 110171-180

66 Loacutepez MD Rosenblad MA Samuelsson T Conserved and variabledomains of RNase MRP RNA RNA Biology 2009 6208-221

67 Marz M Donath A Verstraete N Nguyen VT Stadler PF BensaudeO Evolution of 7SK RNA and its Protein Partners in Meta-zoa Mol Biol Evol 2009 in press

68 Collins LJ Poole AM Penny D Using ancestral sequences touncover potential gene homologues Appl Bioin-formatics 20032(Suppl 3)85-95

69 Chen XS Rozhdestvensky TS Collins LJ Schmitz J Penny D Com-bined experimental and computational approach to identifynon-protein-coding RNAs in the deep-branching eukaryoteGiardia intestinalis Nucleic Acids Res 2007 354619-4628

70 Chen XS White WT Collins LJ Penny D Computational identi-fication of four spliceosomal snRNAs from the deep-branch-ing eukaryote Giardia intestinalis PLoS One 2008 3(8)e3106

71 Barrandon C Spiluttini B Bensaude O Non-coding RNAs regulat-ing the transcriptional machinery Biol Cell 2008 10083-95

72 Hirai H LoVerde PT Identification of the telomeres on Schisto-soma mansoni chromosomes by FISH J Parasitol 199682511-512

73 Theimer CA Feigon J Structure and function of telomeraseRNA Curr Opin Struct Biol 2006 16307-318

74 Stadler PF Chen JJL Hackermuumlller J Hoffmann S Horn F KhaitovichP Kretzschmar AK Mosig A Prohaska SJ Qi X Schutt K Ullmann KEvolution of Vault RNAs Mol Biol Evol 2009 261975-1991

75 Mosig A Zhu L Stadler PF Strategies for Homology-BasedncRNA Gene Annotation Brief Funct Genomics Proteomics 2009 inpress

76 Mosig A Guofeng M Stadler BMR Stadler PF Evolution of theVertebrate Y RNA Cluster Th Biosci 2007 1269-14

77 Perreault J Perreault JP Boire G Ro-associated Y RNAs in meta-zoans evolution and diversification Mol Biol Evol 2007241678-1689

78 Van Horn DJ Eisenberg D OBrien CA Wolin SL Caenorhabditiselegans embryos contain only one major species of Ro RNPRNA 1995 1293-303

79 Boria I Gruber AR Tanzer A Bernhart S Lorenz R Mueller MMHofacker IL Stadler PF Nematode sbRNAs homologs of verte-brate Y RNAs Tech Rep BIOINF-09-020 2009 [httpwwwbioinfuni-leipzigdePublicationsPREPRINTS09-020pdf] Bioinformat-ics University of Leipzig

80 Lartillot N Brinkmann H Philippe H Suppression of long-branchattraction artefacts in the animal phylogeny using a site-het-erogeneous model BMC Evolutionary Biology 2007 7S4

81 Pang KC Stephen S Dinger ME Engstroumlm PG Lenhard B Mattick JSRNAdb 20 -- an expanded database of mammalian non-cod-ing RNAs Nucleic Acids Res 2007 35D178-D182

82 Shin H Hirst M Bainbridge MN Magrini V Mardis E Moerman DGMarra MA Baillie DL Jones SJ Transcriptome analysis forCaenorhabditis elegans based on novel expressed sequencetags BMC Biol 2008 630

83 Inagaki S Numata K Kondo T Tomita M Yasuda K Kanai AKageyama Y Identification and expression analysis of putativemRNA-like non-coding RNA in Drosophila Genes Cells 2005101163-1173

84 Tupy JL Bailey AM Dailey G Evans-Holm M Siebel CW Misra S Cel-niker SE Rubin GM Identification of putative noncoding polya-denylated transcripts in Drosophila melanogaster Proc NatlAcad Sci USA 2005 1025495-5500

85 Zerlotini A Heiges M Wang H Moraes RL Dominitini AJ Ruiz JCKissinger JC Oliveira G SchistoDB a Schistosoma mansonigenome resource Nucleic Acids Res 2009 37D579-D582

86 Liu F Chen P Cui SJ Wang ZQ Han ZG SjTPdb integratedtranscriptome and proteome database and analysis platformfor Schistosoma japonicum BMC Genomics 2008 9304

87 Seemann SE Gilchrist MJ Hofacker IL Stadler PF Gorodkin J Detec-tion of RNA structures in porcine EST data and relatedmammals BMC Genomics 2007 8316

88 Missal K Rose D Stadler PF Non-coding RNAs in Ciona intesti-nalis Bioinformatics 2005 21(S2)i77-i78

89 Missal K Zhu X Rose D Deng W Skogerboslash G Chen R Stadler PFPrediction of Structured Non-Coding RNAs in the Genomeof the Nematode Caenorhabitis elegans J Exp Zool Mol Dev Evol2006 306B379-392

90 Rose DR Hackermuumlller J Washietl S Findeiszlig S Reiche K Hertel JStadler PF Prohaska SJ Computational RNomics of Drosophi-lids BMC Genomics 2007 8406

91 Washietl S Hofacker IL Stadler PF Fast and reliable predictionof noncoding RNAs Proc Natl Acad Sci USA 2005 1022454-2459

92 Hiller M Findeiszlig S Lein S Marz M Nickel C Rose D Schulz C Back-ofen R Prohaska SJ Reuter G Stadler PF Conserved IntronsReveal Novel Transcripts in Drosophila melanogasterGenome Res 2009 191289-1300

93 Lowe T Eddy S tRNAscan-SE a program for improved detec-tion of transfer RNA genes in genomic sequence Nucl AcidsRes 1997 25955-964

94 Griffiths-Jones S Saini HK van Dongen S Enright AJ miRBasetools for microRNA genomics Nucleic Acids Res 200836D154-D158

95 Thompson JD Higgs DG Gibson TJ CLUSTALW improving thesensitivity of progressive multiple sequence alignmentthrough sequence weighting position specific gap penaltiesand weight matrix choice Nucl Acids Res 1994 224673-4680

96 Hofacker IL Fontana W Stadler PF Bonhoeffer LS Tacker M Schus-ter P Fast Folding and Comparison of RNA Secondary Struc-tures Monatsh Chem 1994 125167-188

97 Hofacker IL Fekete M Stadler PF Secondary Structure Predic-tion for Aligned RNA Sequences J Mol Biol 20023191059-1066

98 Lestrade L Weber MJ snoRNA-LBME-db a comprehensivedatabase of human HACA and CD box snoRNAs NucleicAcids Res 2006 34D158-D162

99 Altschul SF Gish W Miller W Myers EW Lipman DJ Basic localalignment search tool J Mol Biol 1990 215403-410

100 Hertel J Hofacker IL Stadler PF snoReport Computationalidentification of snoRNAs with unknown targets Bioinformat-ics 2008 24158-164

101 Lowe TM Eddy SR A Computational Screen for MethylationGuide snoRNAs in Yeast Science 1999 2831168-1171

102 Tafer H Kehr S Hertel J Stadler PF RNAsnoop Efficient targetprediction for box HACA snoRNAs Tech Rep BIOINF-09-0252009 [httpwwwbioinfuni-leipzigdePublicationsPREPRINTS09-025pdf] Bioinformatics University of Leipzig

103 Thompson JD Gibson TJ Plewniak F Jeanmougin F Higgins DG TheCLUSTAL_X windows interface flexible strategies for mul-tiple sequence alignment aided by quality analysis toolsNucleic Acids Res 1997 254876-4882

104 Bernhart SH Tafer H Muumlckstein U Flamm C Stadler PF HofackerIL Partition Function and Base Pairing Probabilities of RNAHeterodimers Algorithms Mol Biol 2006 13

105 Griffiths-Jones S RALEE --RNA ALignment editor in EmacsBioinformatics 2005 21257-259

106 Bailey TL Williams N Misleh C Li WW MEME discovering andanalyzing DNA and protein sequence motifs Nucleic Acids Res2006 34W369-W373

107 Saitou N Nei M The neighbor-joining method a new methodfor reconstructing phylogenetic trees Mol Biol Evol 19874406-425

108 Hofacker IL Vienna RNA secondary structure server NucleicAcids Res 2003 313429-3431

109 Hertel J Lindemeyer M Missal K Fried C Tanzer A Flamm CHofacker IL Stadler PF The Students of Bioinformatics Com-puter Labs 2004 and 2005 The Expansion of the MetazoanMicroRNA Repertoire BMC Genomics 2006 715

Abstract

Background

Results

Conclusion

Background

Results and discussion

Transfer RNAs

Ribosomal RNAs

Spliceosomal RNAs and Spliced Leader RNA

SRP RNA and Ribonuclease P RNA

MicroRNAs

Small Nucleolar RNAs

Other RNA motifs

Uncertain and missing candidates

Conclusion

Methods

tRNA annotation

microRNA annotation

snoRNA annotation

Other RNA families

Additional Data Online

Authors contributions

Additional material

Acknowledgements

References