Gutell 117.rcad_e_science_stockholm_pp15-22


Published on

Ozer S., Doshi K.J., Xu W., and Gutell R.R. (2011).
rCAD: A Novel Database Schema for the Comparative Analysis of RNA.
7th IEEE International Conference on e-Science, Stockholm, Sweden. December 5-8, 2011. pp 15-22.

Published in: Technology
  • Be the first to comment

  • Be the first to like this

No Downloads
Total Views
On Slideshare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Gutell 117.rcad_e_science_stockholm_pp15-22

  1. 1. rCAD: A Novel Database Schema for the Comparative Analysis of RNAStuart Ozer Kishore J. Doshi Weijia Xu Robin R. GutellMicrosoft CorporationCenter for ComputationalBiology and BioinformaticsTexas AdvancedComputing CenterInstitute for Cellular andMolecular Biology1 Microsoft Way The University of Texas at AustinRedmond, WA 98052 Austin, Texas robin.gutell@mail.utexas.eduAbstract-- Beyond its direct involvement in proteinsynthesis with mRNA, tRNA, and rRNA, RNA is nowbeing appreciated for its significance in the overallmetabolism and regulation of the cell. Comparativeanalysis has been very effective in the identification andcharacterization of RNA molecules, including theaccurate prediction of their secondary structure. We aredeveloping an integrative scalable data management andanalysis system, the RNA Comparative AnalysisDatabase (rCAD), implemented with SQL Server tosupport RNA comparative analysis. The platform-agnostic database schema of rCAD captures the essentialrelationships between the different dimensions ofinformation for RNA comparative analysis datasets. TherCAD implementation enables a variety of comparativeanalysis manipulations with multiple integrated datadimensions for advanced RNA comparative analysisworkflows. In this paper, we describe details of therCAD schema design and illustrate its usefulness withtwo usage scenarios.Keywords: Biological Database; RNA Sequence Analysis;Bioinformatics; Database SchemaI. INTRODUCTIONA new perspective is now emerging in Biology: RNAshave a dominant role in the structure, function andregulation of the cell. Like DNA, it has a well-defined set ofrules for nucleotide base pairing – A pairs with U, and Gpairs with C. These consecutive and antiparallel base pairsform canonical helices. Like protein, RNA is capable offorming a three-dimensional structure composed of helices,hairpin, internal, and multi-stem loops, and other structuralmotifs, and like proteins, RNA is capable of catalyzingchemical reactions [1-7]. It is now widely appreciated thatRNA, as a precursor to DNA and proteins, was essential tothe origin of life and to establish the mechanism for theassociation of a cell’s genotype to its phenotype [8-11].Comparative analysis, used effectively by Darwin tocompare and contrast anatomical features of animals [12],has the potential to facilitate the identification andcharacterization of the RNAs primary and higher-orderstructure, and patterns of variation and conservation for theset of analyzed sequences [13]. The utilization ofcomparative analysis is based on a very important discoveryin molecular biology - the same RNA secondary and tertiarystructure can have different RNA sequences [14, 15]. Thepredicted secondary structure models are generallyconserved for each of the specific RNAs. The accuracy fromthese comparative studies is impressive. Approximately98% of the base pairs identified with comparative analysisin the three ribosomal RNAs – 16S, 23S, and 5S thatcontain more than 4,500 nucleotides are in the highresolution crystal structures [16]. In addition to theprediction of an entire RNA structure model, comparativeanalysis has identified RNA structural motifs, the basicbuilding blocks of RNA structure and biases in thedistribution of nucleotides in the ribosomal RNAssecondary structure. These include: unpaired adenosines inthe rRNA secondary structures [17, 18], preponderance oftetraloops - hairpin loops with four nucleotides [19] andother types of irregular structural elements in the ribosomalRNA [20].Successful application of comparative analysis to an RNAdataset requires an interactive workflow that includes theacquisition, management and analysis of large amounts ofbiological information divided into multiple dimensions: 1)sequences and the alignments of those sequences based on acommon structure model, 2) evolutionary relationshipsbetween sequences and 3) higher-order structure andstructural motifs. The iterative nature of the comparativeanalysis workflow ultimately improves the predictedstructure model and the quality of the sequence alignment.However, the comparative analysis workflow does notlend itself to the software pipeline architecture currentlyfavored by most computational biology and bioinformaticsapplications [21, 22]. The pipeline architecture assumes thatrelevant data (e.g., sequence alignment) is primarily storedin flat-files. Different programs load the data from flat-filesinto memory, perform analysis and output the results to adifferent flat-file. The pipeline is created by chainingdifferent programs together. Raw data enters at one end ofthe pipeline and the value-added analyses exit at the otherend in an automated fashion.The comparative analysis workflow requires a semi-automated iterative approach. An important part of thecomparative analysis is the interaction between biologistsand the data. For example, as more sequences becomeavailable, the existing multiple sequence alignment areupdated and curated to improve the comparative model.Similarly, while comparative analysis leads to newdiscoveries about structure and function of RNA sequences,2011 Seventh IEEE International Conference on eScience978-0-7695-4597-4/11 $26.00 © 2011 IEEEDOI 10.1109/eScience.2011.1115
  2. 2. thcaucraosinthpnbdudfeineimaptodsmhese findingscomparative manalysis workfusing non-interWe proposecomparative anrapid increase iand allows a biof RNA datasestructure and evnfrastructure ihe RNA Compersists both thnatural form abetween entitiedatabase manausing a scalabdatabase systemfeatures of theevolving data,ntegration of mThe rCADentities for anmplements aalignment ontphylogenetic) io facilitate andatabase has adsystem eliminminimizes cusubsequentlymodel. Therflow cannot beractive, in-memFigure 1 Overviea different innalysis workfloin size and numiologist to perfets across the dvolutionary relis an efficientmparative Analhe data entitiesand the bioloes. The schemaagement systemble, high perfm, Microsofte rCAD schem, such as chmultiple dimenschema enablRNA datasetprimary objetology: the innformation winalysis and exdvantages overates the transstomized I/Oneed to be intrefore, the fe supported bmory pipeline aew of rCAD systemnfrastructure toow that includember of suitabform manipuladifferent dimenlationships). Adata storage llysis Databasewithin each dogically relevaa is applicable fm and currenformance enteSQL Server 2ma are to effhanges in aligsions of informes direct storin rCAD. Thective of thentegration ofth an RNA seqxploration [23the use of flatslation of daand memotegrated into tfull comparatiy software implement tes support for tble RNA datastion and analynsions (sequencAt the core of olayer (Figuree (rCAD), whiimension in thant relationshifor any relationtly implementerprise relation2008. Two novfficiently suppgnment and tmation.rage of the dahe rCAD systeRNA structustructural (aquence alignme]. A centraliz-files. The rCAta formats, ary managemetheiveolsthetheetsysisce,our1),ichheiripsnaltednalvelorttheataemureandentzedADandentroutinedata. Intwo exaTheanalysiWhile rare relinformaA wRNAcollectirepresemodelsdatabasexternausersalignmetaxonomformatTherFor excontainalignmealignmeprobabitrainedupdatessearchibrowsefeatureanalysisampleproject16S rRNWATpubliclyanalysiincludeexamplread froTheof RNAtypes oforms otheir nrRNAsRNAalignmeRNAmaintaivarioussequencChadsequences. Thus it is mn this paper, wamples of the bII. RrCAD projectis within a cerCAD is a unilated to theation.well-known datfamily databaion of RNA seented by multips. Rfam is prise is frequental data sourcesto browse, sents and comy or keywordfor further anare are other proxample the Rns bacterialents and soments maintaineilistic model sfrom a sets on RDP introing RNA sequer and a taxonoof RDP isis of bacterial res. Another simwhich enableRNA sequencesTERS is a wy available 16is pipeline [22e a dedicatedle, the alignmom external sorCAD projectA sequences anof rRNAs – 5Sof life: bacterinuclear, chlors) as well as osequences areent data is curWeb (CRW)ins other dims analysis feces and their ado is a relatices publishedmore efficientwe describe detabenefits obtainRELATED WOt integrates daentralized relaique applicatiomanagementta source for Rase (Rfam) [2equence familiple sequence alimarily a dataly updated ans. The web insearch and rrresponding cds. Users can dalysis.ojects focusedRibosomal Dand archaealme analysis oed in the RDP psystem. The prof representaoduced new feuence informatomy visualizatits pyrosequerRNA composmilar web sere users to aligns [21].workflow tool6S rDNA ana2]. It doesn’td data managent has to beources into memmaintains a cnd multiple seqS, 16S, and 23Sia, archaea, anroplast, andother dimensioe retrieved drated and avaiSite[13]. Thmensions of ineatures in adalignments.ional databased in recentfor analyzingails of rCAD sned with rCADORKSata curation, aational databason, several othand analysisRNA informa24]. Rfam mies. Each RNAlignments anda provider sernd curated fronterface of Rfaretrieve RNAcovariation mdownload datad on specific RDatabase Projesmall subuof this data[2project are aligrobability paraative sequenceeatures for brotion, includingtion tool. A neencing pipelinition from envrvice is the Gn and analyzethat bundlesalysis tools toaddress data cgement compocomputed onmory for procecomprehensivequence alignmS, from the thrnd eukaryotesmitochondrialons of informdaily from Nlable at the Cohe rCAD schnformation andddition to stoe schema forliterature [26large scaleschema and.access andse system.her projectsof RNAtion is themaintains aA family iscovariancervice. Thisom variousam enablessequencemodels bya in flat fileRNA types.ect (RDP)unit rRNA25]. Thegned with aameters arees. Recentowsing anda genomeew analyticne for thevironmentalGreenGenese their owna suite ofo constructcuration oronent. Forthe fly orssing.e collectionments for allree primary(includingl encodedmation. TheNCBI. Theomparativehema alsod supportsoring rawbiological6]. Chado16
  3. 3. moapemcsaasocaHspmswCrdentorecin(rinoftoasbwRmocebbcabcmanages bioloorganisms withassociated withprotein productexisting ontolomodel organismconform to thissupport genomanalysis. Alignare typically ssequence alignmof the alignmencomparative feanalysis prograHowever, it issequence alignmprovide anmanagement aschema used inwithin the relaChado.III. SCHThe designrequirements: 1dataset (sequeevolutionary renatural relationo the biologirepository supexecution ofcomparative anOur RNA danto three inter(metadata and nrelationships.nformation isontology [23].for different coo be executedavoiding the nseparate analybetween the dawith the biologRNA sequencematrices) as pontology [23].column of theentire alignmenby the databasbase pairs, helicolumns of thealignment are mbe devised tocontaining sequogical knowleh information thh genome sequts encoded byogy and interopm databases ors schema. Theme analysis insments and comstored as pairsment may inclnt. Such schemfeatures and eams (e.g. Blast)less efficientments at largeintegrative pccess and anan rCAD is to fational databaseEMA DESIGN An of rCAD1) persist thences and seqelationships andnships; 2) efficical data, andpporting thecomputationnalysis.ataset for comprrelated dimensnucleotides), stThe organizacongruent witThe unique caomparative anad on very larneed to expoysis applicatioata entities persgical relationse alignments aproposed by tSQL queriese sequence alnt into memoryse. RNA seconices, etc.) aree sequence alimapped onto thretrieve specifuences from spedge for a what can be direuences or the pra genome. Chperability betwr any biologice primary focustead of compmparative featus. For examplude hits and hma is a genericenables results) to be stored ifor storing evscale. The rCAplatform foralysis. Therefoacilitate comme system andAND IMPLEMENis motivatedifferent elemquence alignmd alignments) tiently supportd 3) providemanipulationnal algorithmparative analysisions of informtructure (2-D)ation of ourth the proposedapabilities of Salysis algorithrge RNA dataort large amouons (Figure 1sisted in rCADships and areare stored as 2the RNA strus can be usedignment, withy, using the inndary structuredirectly relategnment. The she Tree of Lifefic fragments opecific taxonomwide varietyectly or indirecrimary RNA ahado is basedween open acceal databases thus of Chado isarative sequenures of sequencle, features ofhigh-scoring pac way for storis from externin a unified wavolving multipAD is designedsequence daore, the databamon analysis tasis different froNTATIONSed by sevements of an RNments, structurthat mimics thfrequent updata central dan and efficiems involvedis is decomposmation: sequenand evolutionadimensionsd RNA structuQL Server alloms and analysasets in-proceunts of data). RelationshiD are coordinatqueried direct2-D grids (sparucture alignmed to access ahout loading tndexing provide elements (e.d to the relevasequences in te, and queries cof the alignmemy subsets.ofctlyandonesshattoncecesf aairsingnalay.pled toataasesksomeralNAres,heirtesataentinsedncearyofureowsesess,toipstedtly.rseentanythededg.,antthecanentThecomparRNA dAlignmand StrA. SequMetaSequenmetadaGenbanSequen16S rRorganelMitochdesigninstancGenbanFigure 2 rCdatabase scrtments extractdataset usedment, Sequenceructure Relationuence MetadataFigure 3adata describinnce Metadataata includenk AccessionnceAccessionRNA) storedllar locationhondrion) storof the Sequece to include renk.CAD database schchema is deted from threefor Comparate Metadata, Evnships (Figureta CompartmenSchema for sequeng RNA sequcompartmentexternal datan and revistable, the sequin Sequencwithin thered in CellLenceAccessionevisions of a shema overview.e-normalizede major dimenstive Analysis:volutionary Ree-2).ntence metadata.uences is stort (Figure 3).abase identifisions) storeduence classificceType tablecell (e.g., NocationInfo ttable allowssequence introinto foursions of anSequenceelationshipsred in theTypes offiers (e.g.,d in theation (e.g.,and theNucleus ortable. Thethe rCADoduced into17
  4. 4. BinaTinPTTtoEnwtas(dCB. EvolutionaryFiguEvolutionaryn the Evolutioand are obtaineThe evolutionan the TaxonomParentTaxID).Taxonomy,TaxonomyNamo link betwEvolutionary Rname of eachwhile all otherable for referestores the full l(e.g., root/celludepth-based quC. Sequence AlFy Relationshipsure 4 Schema for evy relationshipsonary Relationed from the Nary relationshipmy table as a sThe TaxID fiTaxmesOrdered taween the SeRelationshipstaxon is storer names are sences. The Talineage to anyular organisms/ueries of the evlignment Compigure 5 Schema fos Compartmenvolutionary relatio(taxonomy) innships compartNCBI Taxonomps among sequet of parent-chield is the primonomyNamesables and actsequence Metcompartmentsed in Taxonostored in the AaxonomyNameleaf in the Lin/Bacteria) in oolutionary relapartmentor sequence alignmtonshipsrCAD are stortment (Figuremy database [2uences are storhild pairs (TaxImary key for ts aas a foreign kadata and ts. The scientiomyNames tabAlternateNamesOrdered tabneageName fieorder to facilitaationships.mentred4),7].redID,theandkeytheificblemesbleeldateAllSequenmore lfor a spdimensjuxtapoThe juxremovathe aligUnlikesequencalignmeEachuniquetable (stored iof the Anucleotan alignnucleotPhysicaspecificalignmealignmeAlignmTable 15SrRNA16SrRNA23SrRNAThisalignmeupdatesFirst,valuessequencincreasThe ininsertioTree ofin thenucleotFor exarRNA)[13], th(Tablesequencgaps. Tgaps,biological seqnce Alignmentogical sequenpecific RNA msional matrix. Aosed with one axtaposition is aal of gaps. Thugnment matrixin other schemce level, rCAent at nucleotidh sequence aligkey, the AlnID(Figure 5). Min the AlignmAlnID to SeqIDtide in a sequennment as a rowtide isalColumnNumc nucleotide’sent is modifient are idenmentColumn ta: Topology of RibSequences53555762670particular scent has two as when changin, in this schemas they are infces and observses, the numbencreasing sequons and deletiof Life. The resalignment ctides for any seample, for thesequence alighe average per1). Thus, foce alignment,Therefore ourresults in a squences in rCt compartmentnce alignmentsmolecule type cAll sequencesanother to idenaccomplishedus, the contentsx is either a nma where aligAD directly stde level.gnment storedD, and is catalembers of amentSequenceD. The Alignmnce aligned tow record. In Alassociatedber which doerelative positfied. The coluntified andable.bosomal RNA sequTree of LifeTotalAlignmentColumnsMSeLe4049655 317047 5chema designadvantages, spng the alignmema, there is noferred at queryved sequence ver of gaps alsoence variationons observed insult is that thecan greatly eequence withine small subunignment spanninrow ratio of gaor any givenon average 85database schesignificant spaCAD are stort, organized ins. A sequencean be describewithin the aligntify equivalentthrough the ads of any individnucleotide or agnments are stotores multipled in rCAD islogued in the Asequence aligtable throughmentData tablea specific colulignmentDatato aes not changetion within theumns of anymanaged thruence alignments sMax/MinequenceengthAvg.Ratio242/14 73%316/506 85%317/880 85%n for storingace saving anent data.o need to story time. As thevariation in ano increases sign is partially an specific brantotal number oexceed the nn the alignmentit Ribosomal Rng the entire Taps to nucleotirow of the 15% of the coluema, which doace reductionred in thento one oralignmentd as a two-gnment aret positions.ddition anddual cell ina gap (‘-‘).ored at thee sequenceassigned aAlignmentgnment area mappingmaps eachumn withintable, eachspecificunless thate sequencey sequencerough thespanning theGapo (%)% ± 7%% ± 2%% ± 5%sequencend efficientre any gapnumber ofn alignmentgnificantly.a result ofches of theof columnsnumber oft (Table 1).RNA (16STree of Lifeides is 85%16S rRNAumns haveoesn’t storefor storing18
  5. 5. laasLsointhdqoacarFmooPAcmdaothavsarge alignmenand support vsequence alignmFigurSecondly, thLogicalColumnsupports efficioperations onndirection betwhe logical viewdatasets are aquickly becomone entry peralignment incolumnNumberalignment, sucrequire updatinFor example,modified by coonly adds columof where the cPhysicalColumAlignmentColchange, and nmodified (Figudata structurealignments, sucoperations, requhe AlignmentDTo simplifyalignment stovAlignmentGrisequence alignnts and permitvery large (> 6 an example ofhe mapping onNumber in tient (from aa sequence aliween the physiw of the sequadded to rCAes the largest tnucleotide fon an rCADr mapping, anych as insertinng a significanwhen the seqolumn insertionmns to the “enolumn is logicmnNumber to Lumn table is uno rows in thure 6). Withoute, global opech as column iuiring a signifiData table.queries thatred in rCADid and vAlignmment with or wts the rCAD d>106rows xf updating alignmeof PhysicalCothe Alignmendatabase perignment by plical storage inence alignmenD, the Aligntable, even witor each sequenD instance.y further updatg or deletingnt portion of thquence alignmn (Figure 6), tnd” of the aligncally inserted.LogicalColumnupdated to reflhe Alignment the column ierations oninsertions, wouicant number oobtain data fD, two viewmentGridUngawithout gaps rdatabase to sca>104columnent data.olumnNumberntColumn tabrspective) globlacing a layerthe database ant. As new RNmentData tabth a minimumnce within eaWithout thtes of an existia column whe existing rowment topologythe data structunment, regardleThe mappingnNumber in tlect the topolotData table aindirection in tlarge sequenuld be expensiof row updatesfrom a sequenws are createapped that returespectively. Talens)toblebalofandNAbleofachhisingwillws.isureessoftheogyarethenceivetonceed:urnTheabilitysequencadvancstructur7 depicquery fvAlignmsequencamongthe su(Figureonly rerecursivused tocontainalignme(FigureFiguD. Struto retrieve spece alignment thced applicatioral statistics ancts the power ofor retrieving amentGrid -- lce alignmentsequences wiubset are idene 7a). The exaetrieve rowsve query (SQLo identify alln Bacterial seqent is then rete 7c) for the rowure 7 An exemplaructure RelationFigure 8 Secific sub-gridhrough SQL quons of the rnd evolutionarof the rCAD sya subset of a severaging theand the evithin the alignntified by evample query inthat containL Server commrows in the squences (Figuretrieved on a cws that containar SQL query for renships Compartchema of Structurds (rows x coluueries drives mrCAD systemry event countiystem through aequence alignmintegration bevolutionary renment. First, thvolutionary ren Figure 7 is dBacterial seqmon table expsequence aligne 7b). The sucolumn by coln Bacterial seqetrieving partial altmentre Relationshipsumns) of amany of them such asing. Figurean examplement usingetween theelationshipshe rows ofelationshipsdesigned toquences. Apression) isnment thatubset of thelumn basisquences.lignment19
  6. 6. ssbFclooksFTsSeSHtwslo(sesdhlopS4taE4in34The Structustores secondastructure for abase pairs betwFrom the set ofcan be inferredoops (un-paireor multi-stem.The Secondaknown or predsequence alignmFivePrimeElemThreePrimeElesecondary strSecondaryStruelements depeSecondaryStruHelices and inwo extents forsplit into n exteoop. An exten(index), Extentstructural elemextent of the strFigure 9 EThe mappinstructure into tdepicted in Fihelices (5’ 1-4oop (5’ 5-9, 3pairs fromSecondaryStru4, 3’ 32-35) haable (ExtenExtentEndIndex4, 1) and thenternal loop (E3’ half, but the4). The extentural Relationshary structure rgiven sequencween different pf base pairs, otd such as heliceed nucleotides)aryStructureBdicted base paiment. Each basmentSequenceInementSequenceructural elemuctureExtentsending on extuctureExtentTternal loop strr their 5’ and 3ents dependingnt is defined btStartIndex anent is given anructural elemenExample of mappinng of a simplthe Structuraligure 9. The, 3’ 32-35 & 5’ 26-31) and athe two heuctureBasePais two entries inntID, Extenx, ExtentTypeID3’ half is (1,ExtentID 2) alse hairpin loopmodel enablehips compartmrelationships.ce, at a minimpositions of anther structurales (consecutive) classified asBasePairs tablers for any RNse pair is storendexeIndex. The ments are ides table and spltent type enuTypes table.ructural elemen3’ halves. Mulg on the numbeby its first andnd ExtentEndInn identifier, Exnt has its own Eng RNA structure ile stem-loopRelationshipsstem-loop str5’ 10-18, 3’ 20a hairpin loop (elices are eirs table. Then SecondarySntOrdinal, EID) where the 52, 32, 35, 1)so has two extehas only onees SQL queriement (FigureRNA secondamum, is the setn RNA sequencelements/extene base pairs) ahairpin, interne holds the setNA sequence ind with two fielamore complicatentified in tlit into their suumerated in tFor exampnts are split inti-stem loops aer of stems in td last nucleotindex. The entxtentID, and eaExtentOrdinalinto databaseRNA secondacompartmentructure has tw0-25) an intern(16-19). All baentered in tfirst helix (5’tructureExtenExtentStartInde5’ half is (1, 1,(Figure 9). Tents for its 5’ aextent (Extents to characteri8)aryofce.ntsandnalofn aldsandtedtheub-theple,ntoaretheidetireach.aryiswonalasethe1-ntsex,, 1,TheandtIDizethe pattstructurAlgofrom rCotherprogramalreadyexamplpresentcountinA. StrucHerefrequenobservemodelincludeorder sterns of sequenral motif, acrosIV.orithms that opCAD streamlindimensions oms for rCADy organized anles of applicated below: strung.Figure 10 Finctural Statistice, we presentncies. The freed for a base phas many appes the predictiostructure with cnce variation foss the phylogenUSE CASE EXAperate on sequning and manaof data. Thuis not encumnd cross-indexations utilizinguctural statisticnding base pair frecsa common taequency of dipair in a referplications in coon of base paircovariation anfor different elenetic tree.AMPLESuence alignmeaging their accus the develombered with dxed. Two repg the rCAD scs and evolutioequency example.ask of findingifferent baserence secondaromparative anarings in an RNalysis and theements of ants benefitcess to theopment ofdata that ispresentativesystem areonary eventg base pairpair typesry structurealysis. ThisNAs higher-evaluation20
  7. 7. oepcfsscSsd1q(ppaBpthmdsthpnth1pctoHpidathcof the alignmenexisting sequenThe selecteprojected acroscolumn ordinalfrom a referencsequence alignmselected usingcolumn ordinalSQL query issequence alignmdetermine frequ10 bottom).Different struqueries or co(http://www.rnapresentations opotentials forapplication of oB. EvolutionaryFigurPositions inpatterns of varihe RNAs highmethods for iddetermine the fsequences in twhese methodsprediction of anot explicitly dhe evolution o1983 that thepair will enhacovariation anao identify theHowever thephylogenetic trdentified manuand quantify thhat time. Attecomputationalnt accuracy fonce alignment [d base pair,ss the sequencls associated wce row (Figurement includedevolutionaryls for the 5’ adefined to rment, and a gruencies of obsuctural statisticmpiled applica.ccbb.utexas.eon structural sRNA foldinour structural sy Event Countire 11 Evolutionarya sequenceiation (covariather-order strucdentifying thefrequency of eawo columns ofhave been van RNAs highedetermine theof the RNA. Itevolutionary hance the accalysis [31]. Wefirst tertiarynumber ofree (called evoually and thushese phylogeneempts to idenstatistical algor any new seq[13, 28].labeled (1) ine alignment bywith the 5’ ane 10 middle).in the frequenrelationships.and 3’ half ofretrieve nuclerouping statemserved base pacs have been decations. Visitedu/SAE/2D)statistics. Comng programstatistics[29, 30ingy Event Counting ealignment thtion) are usualcture [17, 31]ese positions wach base pair tf the alignmentvery effectiveer-order structunumber of covhas been knowhistory for eacuracy and ree utilized the pinteraction incovariationsolutionary evena rigorous attetic events wantify phylogengorithms of tquence withinn Figure 10,y identifying tnd 3’ nucleotidThe rows of tncy tabulation aUsing the twf the base pairotides from tment is appliedair types (Figueveloped as SQthe CRW Sfor momputing statisticis one speci0].examplehat have similly base paired. The traditionwith covariatitype for all of tt [32, 33]. Whin the accuraure [16], theyvariations duriwn since at lech putative baesolution of tphylogenetic trthe rRNA [3based on tnt counting) wtempt to identas not possibleetic events wthe phylogeneanisthedesthearewo, athetoureQLSiteorecalificilard innalionthehileatedoingastasetheree4].thewasifyatwithetictrees im[35, 36HowprecisechangerCAD.populatcandida11). ThEvolutibasedapplicarecursionodes aversioncurrentmore trfalse pXu, OzPreseapplicapredicaof threesequencsecondaevolutioThetraditiosequencmanipuschemaalignmeUsinperforminsertiosequencevolutiothem dian RNanalysiqueriesbetweetheir ocare easiTheServeralgorithmemornot reqresourcmanagesystemelemenmproved the s].wever, insteade count of the nes on the phylOur methodted with the nuate positions ahe in-memoryionary Relationon the NCation traverseson to deduce tand compute thn of our methot event countrue strong andpositives whenzer, & Gutell, mented here isation of compaated on the intee primary dimce alignmentary, andonary relationsrCAD systemonal computatice alignmentsulated in-memoa directly peents as sparse mng a column indming globalon and deletionce alignmentonary and secirectly to the sNA sequenceis algorithms cs. For exampen sequences, dccurrence on dily determinedrCAD databasenables thehms as compilry space of thequire significance allocationement and inpplaces the rnts on the ensensitivity ofof using statinumber of chanlogenetic treebuilds an inucleotides obtaacross the sequy tree structurnships comparCBI Taxonoms the in-memothe nucleotidehe evolutionaryod was publishting algorithmd weakly covan compared tomanuscript in pV. CONCLUSIOs a new datarative analysiegration and simmensions of infts, associatiothree-dimensiships.m is significaional biologyare primarilyory by differeersists and inmatrices.direction technalignment opn of columnss. The rCADcondary structsequences ande alignment.can be implemeple, structuraldifferent RNAdifferent partsd with implementedevelopmentled applicatione database engnt amounts of cn, parallelnput/output opresponsibilitynterprise datathe covariatioistical methodnges and locatcan be determn-memory treeained from prouence alignmere is obtainedrtment of rCADmy databaseory data struccomposition oy event counts.hed previouslym successfullyariant positionso other methodpreparation).ONStabase schemis to RNA datmultaneous maformation - seqons betweenional structuantly differentworkflows wy stored in flaent programs. Tndexes RNAnique, rCAD isperations inclefficiently forD schema stture entities aindividual nucComparativeented as declarstatistics, reA structural eleof the phylogd on the Micrt of more cons that executegine. These prcustom prograprocessing,timization. Tfor managingabase engine,n methodsds, a moretions of themined withe structure,ojecting theent (Figured from theD, which is[27]. Thecture usingof ancestor. An earliery [37]. Oury identifiess with lessds (Shang,ma for thetasets. It isanipulationquence andprimary,ure, andfrom thewhere RNAat-files andThe rCADsequencecapable ofluding thevery largetores bothand relatescleotides insequencerative SQLelationshipsements andgenetic treerosoft SQLomplicatedwithin therograms doamming formemoryThe rCADg the datawhich is21
  8. 8. optimized for manipulating large quantities of information,and is scalable to support extremely large RNA datasets.Readers who are interested in RNA sequence data canvisit CRW Site ( whichuses rCAD for data management. Readers interested inbuilding customized database can visit for available codes and utilities.ACKNOWLEDGEMENTSThis project has been funded by an External Researchgrant from Microsoft Research, National Institutes of Health(GM067317 and GM085337) and the Welch Foundation(#1427).REFERENCES[1] H. F. Noller and J. B. Chaires, "Functional modification of 16Sribosomal RNA by kethoxal," Proc. Natl. Acad. Sci. USA, vol. 69, pp.3115-8, Nov 1972.[2] K. Kruger, P. J. Grabowski, A. J. Zaug, J. Sands, D. E. Gottschling,and T. R. Cech, "Self-splicing RNA: autoexcision and autocyclizationof the ribosomal RNA intervening sequence of Tetrahymena," Cell,vol. 31, pp. 147-57, Nov 1982.[3] C. Guerrier-Takada, K. Gardiner, T. Marsh, N. Pace, and S. Altman,"The RNA moiety of ribonuclease P is the catalytic subunit of theenzyme," Cell, vol. 35, pp. 849-57, Dec 1983.[4] H. F. Noller, V. Hoffarth, and L. Zimniak, "Unusual resistance ofpeptidyl transferase to protein extraction procedures," Science, vol.256, pp. 1416-9, Jun 5 1992.[5] P. Nissen, J. Hansen, N. Ban, P. B. Moore, and T. A. Steitz, "Thestructural basis of ribosome activity in peptide bond synthesis,"Science, vol. 289, pp. 920-30, Aug 11 2000.[6] B. T. Wimberly, D. E. Brodersen, W. M. Clemons, Jr., R. J. Morgan-Warren, A. P. Carter, C. Vonrhein, T. Hartsch, and V. Ramakrishnan,"Structure of the 30S ribosomal subunit," Nature, vol. 407, pp. 327-39, Sep 21 2000.[7] M. M. Yusupov, G. Z. Yusupova, A. Baucom, K. Lieberman, T. N.Earnest, J. H. Cate, and H. F. Noller, "Crystal structure of theribosome at 5.5 A resolution," Science, vol. 292, pp. 883-96, May 42001.[8] C. R. Woese and G. E. Fox, "The concept of cellular evolution," J.Mol. Evol., vol. 10, pp. 1-6, Sep 20 1977.[9] C. R. Woese, The genetic code; the molecular basis for geneticexpression. New York,: Harper & Row, 1967.[10] L. E. Orgel, "Evolution of the genetic apparatus," J. Mol. Biol., vol.38, pp. 381-93, Dec 1968.[11] F. H. Crick, "The origin of the genetic code," J. Mol. Biol., vol. 38,pp. 367-79, Dec 1968.[12] C. Darwin, The origin of species. New York, NY: Barnes & NobleClassics, 2008.[13] J. J. Cannone, S. Subramanian, M. N. Schnare, J. R. Collett, L. M.DSouza, Y. Du, B. Feng, N. Lin, L. V. Madabusi, K. M. Muller, N.Pande, Z. Shang, N. Yu, and R. R. Gutell, "The comparative RNAweb (CRW) site: an online database of comparative sequence andstructure information for ribosomal, intron, and other RNAs," BMCBioinformatics, vol. 3, p. 2, 2002.[14] R. W. Holley, J. Apgar, G. A. Everett, J. T. Madison, M. Marquisee,S. H. Merrill, J. R. Penswick, and A. Zamir, "Structure of aribonucleic acid," Science, vol. 147, pp. 1462-5, Mar 19 1965.[15] G. E. Fox and C. R. Woese, "5S RNA secondary structure," Nature,vol. 256, pp. 505-7, Aug 7 1975.[16] R. R. Gutell, J. C. Lee, and J. J. Cannone, "The accuracy of ribosomalRNA comparative structure models," Curr. Opin. Struct. Biol., vol.12, pp. 301-10, Jun 2002.[17] R. R. Gutell, B. Weiser, C. R. Woese, and H. F. Noller, "Comparativeanatomy of 16-S-like ribosomal RNA," Prog. Nucleic Acid Res. Mol.Biol., vol. 32, pp. 155-216, 1985.[18] R. R. Gutell, J. J. Cannone, Z. Shang, Y. Du, and M. J. Serra, "Astory: unpaired adenosine bases in ribosomal RNAs," J. Mol. Biol.,vol. 304, pp. 335-54, Dec 1 2000.[19] C. R. Woese, S. Winker, and R. R. Gutell, "Architecture of ribosomalRNA: constraints on the sequence of "tetra-loops"," Proc. Natl. Acad.Sci. USA, vol. 87, pp. 8467-71, Nov 1990.[20] R. R. Gutell, N. Larsen, and C. R. Woese, "Lessons from an evolvingrRNA: 16S and 23S rRNA structures from a comparativeperspective," Microbiol. Rev., vol. 58, pp. 10-26, Mar 1994.[21] T. Z. DeSantis, P. Hugenholtz, N. Larsen, M. Rojas, E. L. Brodie, K.Keller, T. Huber, D. Dalevi, P. Hu, and G. L. Andersen, "Greengenes,a chimera-checked 16S rRNA gene database and workbenchcompatible with ARB," Appl. Environ. Microbiol., vol. 72, pp. 5069-72, Jul 2006.[22] A. L. Hartman, S. Riddle, T. McPhillips, B. Ludascher, and J. A.Eisen, "Introducing W.A.T.E.R.S.: a workflow for the alignment,taxonomy, and ecology of ribosomal sequences," BMCBioinformatics, vol. 11, p. 317, 2010.[23] J. W. Brown, A. Birmingham, P. E. Griffiths, F. Jossinet, R.Kachouri-Lafond, R. Knight, B. F. Lang, N. Leontis, G. Steger, J.Stombaugh, and E. Westhof, "The RNA structure alignmentontology," RNA, vol. 15, pp. 1623-31, Sep 2009.[24] P. P. Gardner, J. Daub, J. G. Tate, E. P. Nawrocki, D. L. Kolbe, S.Lindgreen, A. C. Wilkinson, R. D. Finn, S. Griffiths-Jones, S. R.Eddy, and A. Bateman, "Rfam: updates to the RNA familiesdatabase," Nucleic Acids Res., vol. 37, pp. D136-40, Jan 2009.[25] J. R. Cole, Q. Wang, E. Cardenas, J. Fish, B. Chai, R. J. Farris, A. S.Kulam-Syed-Mohideen, D. M. McGarrell, T. Marsh, G. M. Garrity,and J. M. Tiedje, "The Ribosomal Database Project: improvedalignments and new tools for rRNA analysis," Nucleic Acids Res.,vol. 37, pp. D141-5, Jan 2009.[26] C. J. Mungall and D. B. Emmert, "A Chado case study: an ontology-based modular schema for representing genome-associated biologicalinformation," Bioinformatics, vol. 23, pp. i337-46, Jul 1 2007.[27] D. A. Benson, I. Karsch-Mizrachi, D. J. Lipman, J. Ostell, and E. W.Sayers, "GenBank," Nucleic Acids Res., vol. 37, pp. D26-31, Jan2009.[28] K. J. Doshi, "Ph.D. dissertation. The University of Texas at Austin.,"2007.[29] J. C. Wu, D. P. Gardner, S. Ozer, R. R. Gutell, and P. Ren,"Correlation of RNA secondary structure statistics withthermodynamic stability and applications to folding," J. Mol. Biol.,vol. 391, pp. 769-83, Aug 28 2009.[30] D. P. Gardner, P. Ren, S. Ozer, and R. R. Gutell, "StatisticalPotentials for Hairpin and Internal Loops Improve the Accuracy ofthe Predicted RNA Structure," Journal of Molecular Biology, vol. inpress, 2011.[31] C. R. Woese, R. Gutell, R. Gupta, and H. F. Noller, "Detailedanalysis of the higher-order structure of 16S-like ribosomalribonucleic acids," Microbiol. Rev., vol. 47, pp. 621-69, Dec 1983.[32] D. K. Chiu and T. Kolodziejczak, "Inferring consensus structure fromnucleic acid sequences," Comput. Appl. Biosci., vol. 7, pp. 347-52, Jul1991.[33] R. R. Gutell, A. Power, G. Z. Hertz, E. J. Putz, and G. D. Stormo,"Identifying constraints on the higher-order structure of RNA:continued development and application of comparative sequenceanalysis methods," Nucleic Acids Res., vol. 20, pp. 5785-95, Nov 111992.[34] R. R. Gutell, H. F. Noller, and C. R. Woese, "Higher order structurein ribosomal RNA," EMBO J., vol. 5, pp. 1111-3, May 1986.[35] J. Dutheil, T. Pupko, A. Jean-Marie, and N. Galtier, "A model-basedapproach for detecting coevolving positions in a molecule," Mol. Biol.Evol., vol. 22, pp. 1919-28, Sep 2005.[36] C. H. Yeang, J. F. Darot, H. F. Noller, and D. Haussler, "Detectingthe coevolution of biosequences--an example of RNA interactionprediction," Mol. Biol. Evol., vol. 24, pp. 2119-31, Sep 2007.[37] W. Xu, S. Ozer, and R. R. Gutell, "Covariant Evolutionary EventAnalysis for Base Interaction Prediction Using a Relational DatabaseManagement System for RNA " Lecture Notes in Computer Science,vol. 5566, pp. 200-216, May 2009.22