HindawiPublishingCorporation [619108]

HindawiPublishingCorporation
BioMedResearchInternational
Volume2013,ArticleID901578, 13pages
http://dx.doi.org/10.1155/2013/901578
ReviewArticle
Translational Bioinformatics for Diagnostic and
Prognostic Prediction of Prostate Cancer in the Next-GenerationSequencing Era
Jiajia Chen,1,2Daqing Zhang,1Wenying Yan,1Dongrong Yang,3and Bairong Shen1
1Center for Systems Biology, Soochow University, Suzhou 215006, China
2SchoolofChemistry,BiologyandMaterialEngineering,SuzhouUniversityofScienceandTechnology,Suzhou215011,China
3DepartmentofUrology,TheSecondAffiliatedHospitalofSoochowUniversity,Suzhou215004,China
CorrespondenceshouldbeaddressedtoBairongShen;[anonimizat]
Received1May2013;Accepted22June2013
AcademicEditor:XinghuaLu
Copyright © 2013 JiajiaChenetal. This is an open access article distributed under the Creative Commons Attribution License,
whichpermitsunrestricteduse,distribution,andreproductio ninanymedium,providedtheoriginalworkisproperlycited.
The discovery of prostate cancer biomarkers has been boosted by the advent of next-generation sequencing (NGS) technologies.
Nevertheless,manychallengesstillexistinexploitingthefloodof sequencedataandtranslatingthemintoroutinediagnosticsand
prognosisofprostatecancer.Herewereviewtherecentdevelopmentsinprostatecancerbiomarkersbyhighthroughputsequencing
technologies. We highlight some fundamental issues of translatio n a lb i o i n f o r m a t i c sa n dt h ep o t e n t i a lu s eo fc l o u dc o m p u t i n gi n
NGSdataprocessingfortheimprovementofprostatecancertreatment.
1. Introduction
Prostate cancer (PCa) is the most common cancer and
the second leading cause of cancer deaths among males inwesternsocieties[ 1].Itisestimatedthat241740newPCacases
werediagnosedandthat28170mendiedfromitintheUnitedStates in 2012. Since its discovery over 20 years ago, ProstateSpecific Antigen (PSA) has been the mainstay for diagnosisand prognosis of prostate cancer. However, the routine useof PSA screening remains controversial, owing to its limitedspecificity. PSA fails to differentiate PCa from commonprostatedisorders;moreover,itcannotdiscriminatebetweenaggressive tumors and low-risk ones that may otherwisenever have been diagnosed without screening [ 2]. As such,
overdetection and overtreatment represent critical conse-
quences of PSA-based screening [ 3]. The ongoing debate
highlights the need for more sensitive and specific tools toenablemoreaccuratediagnosisandprognosis.
Duringthelastdecade,theabilitytointerrogateprostate
cancer genomes has rapidly advanced. The resolution forgenomic mutation discovery was improved first with array-based methods and now with next-generation sequencing(NGS) technologies. These high throughput technologiesopen up the possibility to individualize the diagnosis and
treatment of cancer. However significant challenges, partic-ularly with respect to integration, storage, and computationof large-scale sequencing data, will have to be overcome totranslate NGS achievements into the bedside of the cancerpatient. Translational informatics evolves as a promisingmethodologythatcanprovideafoundationforcrossingsuch“translationalbarriers”[ 4,5].
Here we overview the NGS-based strategies in prostate
cancer research, with focus on upcoming biomarker can-didates that show promise for the diagnosis and prognosisof prostate cancer. We also outline future perspectives fortranslational informatics and cloud computation to improveprostatecancermanagement.
2. Microarray Based Diagnosis and
Prognosis of PCa
In the past two decades, high-throughput microarray profil-
inghasbeenutilizedtotrackcomplexmolecularaberrationsduring PCa carcinogenesis. We performed a comprehensives e a r c hi nt h eG e n eE x p r e s s i o nO m n i b u s( G E O )f o rt h e

2 BioMedResearchInternational
Table1:NumberofPCa-associatedGEOseriesgeneratedbymicroarrayandNGS.
MethodologyGeneexpression
profilingNoncodingRNA
profilingGenome
binding/occupancy
profilingGenomemethylation
profilingGenomevariation
profiling
Microarray 266 34 17 21 35
NGS 11 1 18 2 2
array-basedprofilesinhumanPCa.TheretrievedGEOseries
generally fall into 5 categories: gene expression profiling,
noncoding RNA profiling, genome binding/occupancy pro-filing, genome methylation profiling, and genome variationprofiling. The number of GEO series for each category issummarizedin Table 1.
Together these array-based technologies have shed light
on the genetic alterations in the PCa genome. Among theabnormalities affecting prostate tumors, the copy numberalterationisthemostcommonone[ 6].
Numerous early studies have used comparative genome
hybridization (CGH) or single nucleotide polymorphism(SNP)arraystoassesscopynumberchangesintumorDNA.Asaresult,multiplegenomicregionsthatdisplayedfrequentgain or loss in the PCa genome [ 6–11]h a v eb e e nr e v e a l e d .
Chromosome 3p14, 8p22, 10q23, 13q13, and 13q14 are foundto display broad copy number deletion. Key genes mappingwithin these deleted regions include NKX3.1, PTEN [ 12],
BRCA2, C13ORF15, SIAH3 [ 11], RB1, HSD17B2 [ 9], FOXP1,
RYBP, and SHQ1 [ 6]. High-level copy number gains are
detected at 5p13, 14q21, 7q22, Xq12, and 8q13 [ 9]. Key
amplifiedgenesmappingwithintheseregionsincludeSKP2,FOXA1,AR[ 11],andHSD17B3[ 9].
Usingmicroarray,substantialeffortshavealsobeenmade
to characterize prostate cancer gene expression profiles.
Differentiallyexpressedgenesidentifiedinthesestudiespoint
to a plethora of candidate biomarkers with diagnostic orprognosticvalue.
Adiagnosticmarkerisabletodifferentiateprostatecancer
withotherprostaticabnormalities.TherearemanyemergingmarkersthatshowpromiseforPCadiagnosis,suchasalpha-methylacyl-CoA racemase (AMACR) [ 13], prostate cancer
gene 3 (PCA3) [ 14], early prostate cancer antigen (EPCA)-2
[15], Hepsin [ 16], kallikrein-related peptidase 2 (KLK2) [ 17],
and polycomb group protein enhancer of zeste homolog 2(EZH2) [ 18]. The most prominent of these is PCA3, which
was found to exhibit higher sensitivity and specificity forPCa detection than PSA. PCA3 thus provides a potentialcomplementtoPSAfortheearlydiagnosisofPCa.
Prognostic biomarkers for prediction of prostate cancer
patient outcome have also been identified. Increased serumlevels of IL-6 and its receptor (IL-6R) are associated withmetastatic and hormone refractory disease [ 19]. Elevated
serumchromograninAlevelsareindicativeofpoorprogno-sisanddecreasedsurvival[ 20].Otherdifferentiallyexpressed
molecules with prognostic potential include the urokinaseplasminogen activation (uPA) [ 21], TGF-𝛽1[22,23], MUC1
[24],CD24[25],hCAP-D3[ 26],vesicularmonoaminetrans-
porter 2 (SLC18A2) [ 27], TEA domain family member 1(TEAD1),c-Cbl[ 28],SOX7andSOX9[ 29],nuclearreceptor
bindingprotein1(NRBP1)[ 30],CD147[ 31],andWnt5a[ 32].
Eachofthesemarkerswillrequirepropervalidationtoensuretheirclinicalutility.
While microarray technology represents a wonderful
opportunityforthedetectionofgenomicalterations,therearesignificantissuesthatmustbeconsidered.Forexample,array-based methods are impossible to detect variations at a lowfrequency(manywellbelow1%)inthesamples.Inaddition,microarrays can only provide information about the genesthat are already included on the array. The emerging next-generationsequencingtechnology,alsocalledmassivelypar-allel sequencing, however, helps to overcome the challengesbygeneratingactualsequencereads[ 33].
3. NGS Based Diagnosis and Prognosis of PCa
Key feature of the next-generation sequencing technology
is the massive parallelization of the sequencing process. Byvirtue of the massively parallel process, NGS generates hun-dreds of millions of short DNA reads (100–250 nucleotides)simultaneously, which are then assembled and aligned toreferencegenomes.
A number of NGS systems are available commercially,
including Genome Analyzer/HiSeq 2000/MiSeq from Illu-mina, SOLiD/PGM/Proton from Life Sciences, GS-FLX(454)/GSJuniorfromRoche,aswellasnovelsinglemoleculesequencers,forexample,HeliscopefromHelicosBiosciencesandSMRTofferedbyPacificBiosciences.Allthesetechnolo-
giesprovidedigitalinformationonDNAsequencesandmake
it feasible to discover genetic mutations at unprecedentedresolutionandlowercost.
It is generally accepted that cancers are caused by the
accumulationofgenomicalterations.NGSmethodscanwellsqueeze all the alteration information that remains hiddenwithin the genome, including point mutations, small inser-tionsanddeletions(InDels),copynumberalterations(CNV),chromosomal rearrangements, and epigenetic alterations.ForthesereasonsNGShasbecomethemethodofchoiceforlarge-scale detection of somatic cancer genome alterationsandischangingthewayhowcancergenomeisanalyzed.AnNGS-based research pipeline for PCa biomarkers is given inFigure1.
Currently,next-generationsequencingisbeingappliedto
cancergenomestudyinvariousways:
(1) genome-based sequencing (DNA-Seq), yielding in-
formation on sequence variation, InDels, chromoso-malrearrangements,andcopynumbervariations,

BioMedResearchInternational 3
Next-generation sequencing technologies
DNA-Seq Chip-Seq RNA-Seq Methyl-Seq
CNV Chromosomal
rearrangement
Point mutationAlternative splicing Gene fusionTFBS
DNA methylation
Histone modificationRNA expression and discovery
Pathway databases
Statistical methods
Regulatory network
Cancer-related pathwaysMolecular
alteration
Literatures
id d i
11
22
345 67 8 9 10 11 12 13 14
G U AmRNA
5󳰀CG
GCC
GG
C
MMM
M
……
Zfp198 Elys
Err2
Rif1
BAF155Wdr18Btbd14aArid3aTif1BSall1Pelo
Zfp609
MybbpNF45SP1
HDAC2Cdk1
EWS
Zfp219
Rybp
RequiemArid3bWapl
P66B
Rex1Oct4Zfp281Nanog Dax1
Nac1Sall4REST
Rnf2Rai14
Prmt1YY1
MAPK
Figure1:NGS-basedpipelineforcancermarkerdiscovery.
(2) transcriptome-based sequencing (RNA-Seq), yield-
ing quantitative information on transcribed regions(totalRNA,mRNA,ornoncodingRNAs),
(3) interactome-based sequencing (ChIP-Seq), yielding
information on protein binding sequences and his-tonemodification,
(4) methylome-based sequencing (Methl-seq), yielding
quantitative information on DNA methylation andchromatinconformation.
Inthefollowingpart,wewillintroducethemainapplica-
tionstonext-generationsequencingofprostatecancer,usingexamplesfromtherecentscientificliterature(summarizedinTable 2).
3.1. DNA-Seq. According to the proportion of the genome
targeted, DNA-Seq is categorized into whole-genome seq-uencing and exome sequencing. The goal of whole-genomesequencing is to sequence the entire genome, not just cod-ing genes, at a single-base resolution. By whole-genomesequencing, recent studies have provided detailed landscapeof genomic alterations in localized prostate cancers [ 6,11,
46,68,69]. The full range of genomic alterations that drive
prostatecancerdevelopmentandprogression,includingcopynumbergainsandlosses,singlenucleotidesubstitutions,andchromosomalrearrangements,arereadilyidentified.
3.1.1.CopyNumberAlteration. Mostprostatecancersexhibit
somatic copy number alterations, with genomic deletionsoutnumbering amplifications [ 6]. Early methods for copy
number analysis involve fluorescence in situ hybridizationsand array-based methods (CGH arrays and SNP arrays).Morerecently,NGStechnologieshavebeenutilizedandoffersubstantialbenefitsforcopynumberanalysis.
NGS used changes in sequencing depth (relative to a
normalcontrol)toidentifycopynumberchanges.Thedigital
nature of NGS therefore allows accurate estimation of copy
number levels at higher resolution. In addition, NGS canprovide novel gene copy information such as homozygousandheterozygousdeletionsandgeneamplifications,whereastraditional sequencing approaches cannot. For example,by next-generation sequencing of castrate-resistant prostatecancer (CRPC), Collins et al. [ 34] identified a homozygous
9p21 deletion spanning the MTAP, CDKN2, and ARF genesand deficiency of MTAP was suggested as an exploitabletumortarget.
3.1.2.SomaticNucleotideSubstitutions. Whilewhole-genome
sequencing provides the most comprehensive characteriza-tionofthecancergenome,itisthemostcostly.Alternatively,targeted sequencing approaches, such as exome sequencing,
assemble multiple cancer genomes for analysis in a cost-
effective manner. Whole-exome sequencing captures thecodingexonsofgenesthatcontainthevastmajorityofdiseasecausing mutations. Relative to structural alterations, pointmutationsarelesscommoninprostatecancer[ 6,70]andthe
average mutation rate was estimated at 1.4Mb
−1in localized
PCa[35]and2.0Mb−1inCRPC[ 36].

4 BioMedResearchInternational
Table2:SummaryofNGS-basedstudiesonprostatecancer.
Discoveries Method References
CopynumberlossofMTAP,CDKN2,andARFgenes
DNA-Seq[34]
SomaticmutationsinMTOR,BRCA2,ARHGEF12,andCHD5genes [ 11]
NCOA2,p300,theARcorepressorNRIP1/RIP140,andNCOR2/SMRT [ 6]
SomaticmutationsinSPOP,FOXA1,andMED12 [35]
SomaticmutationsinMLL2andFOXA1 [36]
SomaticmutationsinTP53,DLK2,GPC6,andSDF4. [ 37]
TMPRSS2:ERG,TMPRSS2:ETV1
RNA-Seq[38]
TMPRSS2:ETV4 [39]
TMPRSS2:ETV5,SLC45A3:ETV5 [40]
TMPRSS2:ELK4 [41]
SLC45A3:ETV1,HERV-K 22q11.23:ETV1,HNRPA2B1:ETV1,andC15ORF21:ETV1 [ 42]
KLK2:ETV4andCANT1:ETV4 [43]
SLC45A3:BRAForESRP1:RAF1 [44]
C15orf21:Myc [45]
EPB41:BRAF [46]
TMEM79:SMG5 [47]
DifferentialexpressionofPCAT-1 [48]
DifferentialexpressionofmiR-16,miR-34a,miR-126∗,miR-145,andmiR-205 [ 49]
HDACsandEZH2workasERGcorepressors
Chip-Seq[50]
AP4asanovelco-TFofAR [51]
POU2F1andNKX3-1 [52]
Runx2aregulatessecretioninvasivenessandmembranesecretion [ 53]
AnoveltranscriptionalregulatorynetworkbetweenNKX3-1,AR,andtheRABGTPasesignalingpathway [ 54]
Distinctpatternsofpromotermethylationaroundtranscriptionstartsites Methyl-Seq [ 55]
Capillary-based exome sequencing has extensively been
performed in localized PCa and CRPC, and a handful ofoncogenic point mutations have been defined. Remarkably,Taylor et al. [ 6]performedfocusedexonresequencingin218
prostatecancertumorsandidentifiedmultiplesomaticalter-ations in the androgen receptor (AR) gene as well as itsupstream regulators and downstream targets. For example,t h eA Rc o a c t i v a t o rN C O A 2a n dp 3 0 0 ,t h eA Rc o r e p r e s s o rNRIP1/RIP140 and NCOR2/SMRT were found to harbor
somatic mutations. Other genes including KLF6, TP53, AR,
EPHB2, CHEK2, and ATBF1 [ 6,71–74]h a v ea l s ob e e nr e –
portedtoharborsomaticmutationsinlocalizedprostatecan-cer.
RecentlyNGSisbecomingincreasinglyroutineforexome
sequencing analysis. Whole-exome sequencing using next-generationsequencing(NGS)technologiescomparesallexonsequences between tumors and matched normal samples.Multiplereadsthatshownonreferencesequencearedetectedaspointmutations.Inthiswayanumberofdrivermutationsin prostate cancer have been uncovered. Robbins et al. [ 11]
used NGS-based exome sequencing in 8 metastatic prostatetumorsandr evealedno velsoma ticpoin tm uta tionsingenesincluding MTOR, BRCA2, ARHGEF12, and CHD5. Kumaret al. [37] performed whole-exome sequencing of lethal
metastatictumorsandhigh-gradeprimarycarcinomas.TheyalsoobservedsomaticmutationsinTP53,DLK2,GPC6,andSDF4.MorerecentlyBarbierietal.[ 35]andGrassoetal.[ 36]
systematicallyanalyzedsomaticmutationsinlargecohortsofprostate tumors. Barbieri et al. [ 35] investigated 112 primary
tumor-normal pairs and revealed novel recurrent mutationsi nS P O P ,F O X A 1 ,a n dM E D 1 2 .G r a s s oe ta l .[ 36] sequenced
the exomes of 11 treatment-naive and 50 lethal CRPC andidentified recurrent mutations in multiple chromatin- andhistone-modifyinggenes,includingMLL2andFOXA1.Thesetwo studies also reported mutated genes (SPOP [ 35]a n d
CHD1 [36]) that may define prostate cancer subtypes which
areETSgenefamilyfusionnegative.
Together these findings present a comprehensive list of
specific genes that might be involved in prostate cancer andprioritizecandidatesforfuturestudy.
3.2.RNA-Seq. Inadditiontogenomeapplications,NGSwill
also dramatically enhance our ability to analyze transcrip-
tomes. Before NGS, microarrays have been the dominanttechnologyfortranscriptomeanalysis.Microarraytechnolo-gies rely on sequence-specific probe hybridization, and flu-orescence detection to measure gene expression levels. It issubject to high noise levels, cross-hybridization and limiteddynamic range. Compared to microarrays, the emergingRNA-Seqprovidesdigitalgeneexpressionmeasurementsthatoffersignificantadvantagesinresolution,dynamicrange,andreproducibility. The goal of transcriptome sequencing is tosequence all transcribed genes, including both coding andnoncoding RNAs. It is independent of prior knowledge andoffers capacity to identify novel transcripts and mutations

BioMedResearchInternational 5
that microarrays could not achieve, such as fusion genes,
noncodingRNAs,andsplicevariants.
3.2.1. Gene Fusions. Recurrent gene fusion is a prevalent
typeofmutationresultingfromthechromosomalrearrange-ments, which can generate novel functional transcripts thatserve as therapeutic targets. Early studies relied on cytoge-neticmethodstodetectchromosomalrearrangements.How-everthismethodisonlyapplicableincasesofsimplegenomesa n di sv u l n e r a b l ei nc o m p l e xg e n o m e so fe p i t h e l i a lc a n c e r ssuchasPCa.
Complete sequencing of prostate cancer genomes has
provided further insight into chromosomal rearrangementsin prostate cancer. NGS technologies, for example, paired-endsequencingapproachesaresufficientlysensitivetodetectbreakpointcrossingreadsandareextremelypowerfulforthediscoveryoffusiontranscriptsandpotentialbreakpoints.
The first major recurrent fusion to be identified in
prostatecancerwasdiscoveredbyTomlinsetal.usingCancerOutlier Profile Analysis (COPA) algorithm [ 38]. The fusion
discovered places two oncogenic transcription factors fromthe ETS family (ETV1 and ERG) under control of theprostate-specificgeneTMPRSS2.
While the TMPRSS2:ETV1 fusion is rare and occurs in
1–10% of prostate cancers [ 75], the TMPRSS2:ERG fusion is
present in roughly half of prostate cancers and is the most
common genetic aberration so far described in prostate
cancer. Furthermore TMPRSS2:ERG is unique only to pro-static nonbenign cancers [ 76]. Given this high specificity, its
clinical application as ancillary diagnostic test or prognosticbiomarker is promising. The expression of TMPRSS2:ERGfusion gene has been proposed as a diagnostic tool, alone orin combination with PCA3 [ 77]. In addition, many studies
have suggested that TMRSS2:ERG could be a prognosticbiomarkerforaggressiveprostatecancer.
Following Tomlins’ pioneering discovery, subsequent
research has identified a host of similar ETS family genefusions. Other oncogenic ETS transcription factors, forexample, ETV4 [ 39], ETV5 [ 40], and ELK4 [ 41], have been
identifiedasadditionalfusionpartnersforTMPRSS2.Otherunique 5
󸀠fusion partner genes to ETS family members have
also been identified, such as SLC45A3, HERV-K 22q11.23,
HNRPA2B1,andC15ORF21infusionwithETV1[ 42],KLK2
and CANT1 in fusion with ETV4 [ 43], and SLC45A3:ETV5
[40].
For ETS fusion-negative prostate cancers subtypes,
novel gene fusions have also been identified, includingSLC45A3:BRAF, ESRP1:RAF1, SLC45A3:BRAF, ESRP1:RAF1[44], C15orf21:Myc [ 45], EPB41:BRAF [ 46], and
TMEM79:SMG5[ 47].
3.2.2. Noncoding RNAs. In addition to gene fusions, RNA-
Seq also enables discovery of new noncoding RNAs (ncR-NAs)withthepotentialtoserveascancermarkers.Transcrip-tomesequencingofaprostatecancercohorthasidentifiedanunannotated ncRNA PCAT-1 as a transcriptional repressorlinked to PCa progression [ 48]. RNA- Seq was also applied
toidentifydifferentiallyexpressedmicroRNAs(e.g., miR-16,miR-34a, miR-126
∗, miR-145, and miR-205) associated withmetastatic prostate cancer [ 49]. These findings establish the
utilityofRNA-Seqtoidentifydisease-associatedncRNAsthatcouldprovidepotentialbiomarkersortherapeutictargets.
3.3.ChIP-Seq. AnotherapplicationofNGScapitalizesonthe
abilitytoanalyzeprotein-DNAinteractions,asforChIP-Seq.ChIP-Seq provides clear indications of transcription factorbindingsites(TFBSs)athighresolution.Itisalsowellsuitedfordetectingpatternsofmodifiedhistonesinagenome-widemanner[78].
Much of gene regulation occurs at the level of transcrip-
tional control. In addition, aberrant histone modifications(methylation or acetylation) are also associated with cancer.Therefore experimental identification of TFBSs or histonemodificationshasbeenanareaofhighinterest.IntraditionalChIP-chip approaches, DNA associated with a transcriptionfactor or histone modification of interest is first selectivelyenriched by chromatin immunoprecipitation, followed byprobing on DNA microarrays. In contrast to ChIP-chip,ChIP-Seq uses NGS instead of custom-designed arrays toidentify precipitated DNA fragments, thus yielding moreunbiasedandsensitiveinformationabouttargetregions.
Many efforts have employed ChIP-Seq approaches to
characterize transcriptional occupancy of AR, and manynoveltranslationalpartnersofARhavebeenidentified.UsingChIP-Seq, Chng et al. [ 50] performed global analysis of AR
a n dE R Gb i n d i n gs i t e s .Th e yr e v e a l e dt h a tE R Gp r o m o t e s
prostate cancer progression by working together with tran-scriptionalcorepressorsincludingHDACsandEZH2.Zhanget al. [51]d e v e l o p e dac o m o t i fs c a n n i n gp r o g r a mc a l l e d
CENTDIST and applied it on an AR ChIP-Seq dataset fromaprostatecancercellline.Theycorrectlypredictedallknownco-TFs of AR as well as discovered AP4 as a novel AR co-
TF. Little et al. [ 53] used genome-wide ChIP-Seq to study
Runx2 occupancy in prostate cancer cells. They suggestednovelroleofRunx2ainregulatingsecretioninvasivenessandmembrane secretion. Tan et al. [ 54] showed that NKX3-1
colocalizes with AR and proposed a critical transcriptionalregulatory network between NKX3-1, AR, and the RABGTPase signaling pathway in prostate cancer. Urbanucci etal. [79] identified AR-binding sites and demonstrated that
the overexpression of AR enhances the receptor binding tochromatin in CRPC. By comparing nucleosome occupancymaps using nucleosome-resolution H3K4me2 ChIP-Seq, Heet al. [52] found that nucleosome occupancy changes can
predict transcription factor cistromes. This approach alsocorrectly predicted the binding of two factors, POU2F1 andNKX3-1.Byhigh-resolutionmappingofintra-andinterchro-mosomeinteractions,Rickmanetal.[ 80]demonstratedthat
ERGbindingisenrichedinhotspotsofdifferentialchromatin
interaction.TheirresultindicatedthatERGoverexpressionis
capableofinducingchangesinchromatinstructures.
Taken together, these studies have provided a complete
regulatorylandscapeinprostatecancer.
3.4.Methyl-Seq. AnotherfertileareaforNGSinvolvesassess-
mentofthegenome-widemethylationstatusofDNA.Methy-lation of cytosine residues in DNA is known to silenceparts of the genome by inducing chromatin condensation.

6 BioMedResearchInternational
DNA hypermethylation probably remains most stable and
abundantepigeneticmarker.
Several DNA methylation markers have been identified
in prostate cancer. The most extensively studied one isCpG island hypermethylation of glutathione-S-transferaseP (GSTP1) promoter DNA, resulting in the loss of GSTP1expression [ 81]. Today GSTP1 hypermethylation is most fre-
quentlyevaluatedasdiagnosticbiomarkerforprostatecancer.Itisalsoanadverseprognosticmarkerthatpredictsrelapseofpatientsfollowingradicalprostatectomy[ 82].
WiththeaidofNGStechnologies,genome-widemapping
of methylated cytosine patterns in cancer cells becomefeasible.Basedonestablishedepigeneticsmethods,thereareemerging next-generation sequencing applications for theinterrogation of methylation patterns, including methyl-ation-dependent immunoprecipitation sequencing (MeDIP-Seq) [83], cytosine methylome sequencing (MethylC-Seq)
[84],reducedrepresentationbisulfitesequencing(RRBS-Seq)
[85], methyl-binding protein sequencing (MBP-Seq) [ 86],
andmethylationsequencing(Methyl-Seq)[ 87].Anumberof
aberrantmethylationprofileshavebeendevelopedsofarandare being evaluated as potential markers for early diagnosisand risk assessment. As an example, using MethylPlex-Seq,
Kimetal.[ 55]mappedtheglobalDN Amethylationpatterns
inprostatetissuesandrevealeddistinctpatternsofpromotermethylation around transcription start sites. The compre-hensive methylome map will further our understanding ofepigeneticregulationinprostatecancerprogression.
4. PCa Biomarkers in Combinations
Diagnostic and prognostic markers with relevance to PCaare routinely identified. Nevertheless, we should stay awarethat candidate markers obtained are often irreproduciblefrom experiment to experiment and very few molecules willmakeittotheroutineclinicalpractice.Cancerisanonlineardynamic system that involves the interaction of many bio-logicalcomponentsandisnotdrivenbyindividualcausativemutations.Inmostcases,nosinglebiomarkerislikelytodic-tatediagnosisorprognosissuccess.Consequently,thefutureof cancer diagnosis and prognosis might rely on the combi-nationofapanelofmarkers.Kattanandassociates[ 88]have
establishedaprognosticmodelthatincorporatesserumTGF-𝛽1 and IL-6R for prediction of recurrence following radical
prostatectomy. This combination shows increased predictiveaccuracy from 75 to 84%. Furthermore, a predictive modelincorporating GSTP1, retinoic acid receptor 𝛽2( R A R𝛽2),
andadenomatouspolyposiscoli(APC)hasbeenassessed.ButnoincreaseddiagnosticaccuracywasshowncomparedwithPSAlevelalone[ 89].Morerecentlyapanelofmarkers,PCA3,
serum PSA level, and % free PSA show improved predictivevalue compared with PSA level and % free PSA. Again theclinicalutilityofthesecombinationsneedstobeevaluatedinlarge-scalestudies.
5. PCa Biomarkers at Pathway Level
Recently,animportanceofpathwayanalysishasbeenempha-sized in the study of cancer biomarkers. Pathway-basedapproach allows biologists to detect modest expressionchangesoffunctionallyimportantgenesthatwouldbemissed
in expression-alone analysis. In addition, this approachenables the incorporation of previously acquired biologicalknowledge and makes a more biology-driven analysis of ge-nomicsdata.Pathwayanalysistypicallycorrelatesagivensetofmolecularchanges(e.g.,differentialexpression,mutation,
and copy number variation data) by projecting them onto
well-characterizedbiologicalpathways.Anumberofcurateddatabasesareavailableforcanonicalsignalingandmetabolicpathways,suchasKyotoEncyclopediaofGenesandGenom-es (KEGG, http://www.genome.jp/kegg/ ), Molecular Signa-
tures Database (MSigDB), IngenuityPathway Analysis (IPA,http://www.ingenuity.com/ ), GeneGO by MetaCore ( http://
www.genego.com/ ) ,a n dG e n eS e tE n r i c h m e n tA n a l y s i s
(GSEA,http://www.broadinstitute.org/gsea/ ).Enrichmentof
pathways can be evaluated by overrepresentation statis-tics. The overall flowchart of the proposed pathway-basedbiomarker approach is illustrated in Figure1.U s i n gt h i s
pipeline, the pathways enriched with aberrations are identi-fiedandthenproposedtobethepotentialcandidatemarkers.
Some impressive progress has been made to identify
pathways with relevance to the pathophysiology of prostatecancer.Rhodesetal.[ 90]wereofthefirsttoperformpathway
analysis of the microarray expression datasets. By meta-analysisof4independentmicroarraydatasets,theygenerateda cohort of genes that were commonly deregulated in PCa.Theauthorsthenmappedtheidentifiedderegulatedgenestofunctionalannotationsandpinpointedpolyamineandpurinebiosynthesis as critical pathway altered in PCa. Activationof Wnt signaling pathway was reported to be key pathways
defining the poor PCa outcome group [ 91]. A comparison
ofcastration-resistantandcastration-sensitivepairsoftumorlineshighlightedtheWntpathwayaspotentiallycontributingto castration resistance [ 37]. Using a similar pathway-based
approach,Wangetal.foundEndothelin-1/EDNRAtransacti-vationoftheEGFRaputativenovelPCarelatedpathway[ 92].
More recently, an integrative analysis of genomic changesrevealed the role of the PI3K, RAS/RAF, and AR pathwaysinmetastaticprostatecancers[ 6].Theaboveinsightsprovide
a blueprint for the design of novel pathway inhibitors intargetedtherapiesforprostatecancer.
Thusfarthepathway-basedapproachholdsgreatpromise
for cancer prediction. However, the known pathways corre-spond merely to a small fraction of somatic alterations. Thealterationsthathavenotbeenassignedtoadefinitivepathwayundermine the basis for a strictly pathway-centric markerdiscovery. In addition, the cross-talk between differentsignaling pathways further complicated the pathway-basedanalysis. Thus, biomarker discovery has to shift toward anintegrative network-based approach that accounts moreextensivegenomicalteration.
6. PCa-Specific Databases
High throughput research in PCa has led to vast amountsof comprehensive datasets. There has been a growing desireto integrate specific data types into a centralized databaseand make them publicly available. Considerable efforts wereundertaken and to date various national, multicenter, and

BioMedResearchInternational 7
institutional databases in the context of prostate cancer re-
search are available. Prostate gene database (PGDB, http://
www.urogene.org/pgdb/ )i sac u r a t e dd a t a b a s eo ng e n e so r
genomiclocirelatedtohumanprostateandprostaticdiseases[93].Anotherdatabase,prostateexpressiondatabase(PEDB,
http://www.pedb.org/ )i sac u r a t e dd a t a b a s et h a tc o n t a i n s
toolsforanalyzingprostategeneexpressioninbothcancerous
and normal conditions [ 94]. Dragon database of genes
associated with prostate cancer (DDPC, http://cbrc.kaust
.edu.sa/ddpc/ )[95]isanintegratedknowledgedatabasethat
providesamultitudeofinformationrelatedtoPCaandPCa-related genes. ChromSorter [ 96]c o l l e c t sP C ac h r o m o s o m a l
regions associated with human prostate cancer. PCaMDB isa genotype-phenotype database that collects prostate cancerrelated gene and protein variants from published literatures.These specific databases tend to include large numbers ofpatients from different geographic regions. Their general-izability and statistical power offer researchers a uniqueopportunity to conduct prostate cancer research in variousareas.
7. Translational Bioinformatics in PCa:
A Future Direction
Recent advances in NGS technologies have resulted in huge
sequencedatasets.Thisposesatremendouschallengefortheemerging field of translational bioinformatics. Translationalbioinformatics, by definition, is the development of storage,analytic, and interpretive methods to optimize the transla-tion from bench (laboratory-based genomic discoveries) tobedside (evidence-based clinical applications). The aim oftranslational bioinformatics is to combine the innovationsand resources across the entire spectrum of translationalmedicinetowardsthebettermentofhumanhealth.Toachievethis goal, the fundamental aspects of bioinformatics (e.g.,bioinformatics,imaginginformatics,clinicalinformatics,andpublichealthinformatics)needtobeintegrated( Figure2).
As the first aspect, bioinformatics is concerned with
applyingcomputationalapproachestocomprehendtheintri-catebiologicaldetailselucidatingmolecularandcellularpro-cessesofcancer.Imaginginformaticsisfocusedonwhathap-pens at the level of tissues and organs, and informatics tech-niques are used for image interpretation. Clinical bioinfor-maticsfocusesondatafromindividualpatients.Itisorientedtoprovidethetechnicalinfrastructuretounderstandclinical
risk factors and differential response to treatment at the
individual levels. As for public health informatics, the strat-ified population of patients is at the center of interest. Infor-matics solutions are required to study shared genetic, envi-ronmental,andlifestyleriskfactorsonapopulationlevel.Inorder to achieve the technical and semantic interoperabilityofmultidimensionaldata,fundamentalissuesininformationexchangeandrepositoryaretobeaddressed.
8. Future Perspectives
The prospects for NGS-based biomarkers are excellent.However, compared with array-based studies, real in-depthNGS is still costly and also the analysis pipeline is less
established, thus challenging the use of NGS. A survey inthe GEO indicated that NGS is currently finding modestapplication in the identification of PCa markers. Table 1
l i s t e dt h en u m b e ro fp u b l i s h e dG E Os e r i e so nh u m a nP C ausingNGSversusmicroarrays.Therefore,furtherworkisstill
requiredbeforeNGScanberoutinelyusedintheclinic.Issues
regardingdatamanagement,dataintegration,andbiologicalvariationwillhavetobetackled.
8.1. NGS in the Cloud. The dramatic increase in sequencer
output has outpaced the improvements in computationalinfrastructurenecessarytoprocessthehugevolumesofdata.Fortunately, an alternative computing architecture, cloudcomputing, has recently emerged, which provides fast andcost-effectivesolutionstotheanalysisoflarge-scalesequencedata. In cloud computing, high parallel tasks are run ona computer cluster through a virtual operating system (or“cloud”). Underlying the clouds are compact and virtual-ized virtual machines (VMs) hosting computation-intensiveapplicationsfromdistributedusers.Cloudcomputingallowsusersto“rent”processingpowerandstoragevirtuallyontheirdemand and pay for what they use. There has been con-siderable enthusiasm in the bioinformatics community foruseofcloudcomputingservices. Figure3showsaschematic
drawingofthecloud-basedNGSanalysis.
Recently exploratory efforts have been made in cloud-
based DNA sequence storage. O’Connor et al. [ 97]c r e a t e d
SeqWareQueryEngineusingcloudcomputingtechnologiesto support databasing and query of information from thou-sands of genomes. BaseSpace is a scalable cloud-computingplatform for all of Illumina’s sequencing systems. After asequencing run is completed, data from sequencing instru-
mentsisautomaticallyuploadedtoBaseSpaceforanalysisand
storage. DNAnexus, a company specialized in Cloud-based
DNA data analysis, has leveraged the storage capacity ofGoogleCloudtoprovidehigh-performancestorageforNGS
data.
Some initiatives have utilized preconfigured software
on such cloud systems to process and analyze NGS data.Table 3summarizessometoolsthatarecurrentlyavailablefor
sequence alignment, short read mapping, single nucleotidepolymorphism (SNP) identification, and RNA expression
analysis,amongstothers.
Although cloud computing seems quite attractive, there
arealsoissuesthatareyettoberesolved.Themostsignificant
concernspertaintoinformationsecurityandbandwidthlim-
itation. Transferring massive amounts of data (on the orderof petabytes) to the cloud may be time consuming and
prohibitively expensive. For most sequencing centers that
require substantial data movement on a regular basis, cloud
computingcurrentlydoesnotmakeeconomicsense.Whileit
isclearthatcloudcomputinghasgreatpotentialforresearchpurposes,forsmalllabsandclinicalapplicationsusingbench-
top genome sequencers of limited throughput, like MiSeq,
PGM,andProton,cloudcomputingdoeshavesomepractical
utility.

8 BioMedResearchInternational
Bioinformatics
Molecules
Translational
biomedical
knowledgebase
Imaging informatics
Tissues and organs
Clinical informatics
IndividualsPublic health informatics
Population/society
Figure2:Translationalbioinformaticsbridgesknowledgefrommolecules topopulations.Foursubdisciplinesoftranslationalbioinformatics
andtheirrespectivefocusareasaredepictedinboxes.Thesuccessoftranslationalbioinformaticswillenableacompleteinformationlogistics
chainfromsinglemoleculestotheentirehumanpopulationandthuslinkinnovationsfrombenchtobedside.
Cloud-based web services
Data storageMaster virtual machineWorker nodes
ResearcherSequencing lab
Job queue
Data partitionData
Figure3:Schematicofthecloud-basedNGSanalysis.Localcomputersallocatethecloud-basedwebservicesovertheinternet.Webservices
comprised a cluster of virtual machines (one master node and a chosen number of worker nodes). Input data are transferred to the cloud
storage and the program code driving the computation is uploaded to master nodes, by which worker nodes are provisioned. Each worker
nodedownloadsreadsfromthestorageandruncomputationindependently.Thefinalresultisstoredandmeanwhiletransferredtothelocalclientcomputerandthejobcompletes.

BioMedResearchInternational 9
Table3:ThecloudcomputingsoftwareforNGSdataanalysis.
Software Website Description References
Crossbow http://bowtie-bio.sourceforge.net/crossbow/ ReadmappingandSNPcalling [56]
CloudBurst http://cloudburst-bio.sourceforge.net/ Reference-basedreadmapping [57]
Contrail http://contrail-bio.sourceforge.net/ Denovoreadassembly [58]
Cloud-MAQ http://sourceforge.net/projects/cloud-maq/ Readmappingandassembly [59]
Bioscope http://www.lifescopecloud.com/ Reference-basedreadmapping [60]
GeneSifter http://www.geospiza.com/Products/AnalysisEdition.shtmlCustomerorientedNGSdataanalysis
services[61]
CloudAligner http://sourceforge.net/projects/cloudaligner/ Readmapping [62]
Roundup http://rodeo.med.harvard.edu/tools/roundupOptimizedcomputationforcomparative
genomics[63]
PeakRanger http://www.modencode.org/software/ranger/ PeakcallerforChIP-Seqdata [64]
Myrna http://bowtie-bio.sf.net/myrna/Differentialexpressionanalysisfor
RNA-Seqdata[65]
ArrayExpressHTS http://www.ebi.ac.uk/Tools/rwiki/RNA-Seqdataprocessingandquality
assessment[66]
SeqMapreduce Notavailable Readmapping [67]
BaseSpace https://basespace.illumina.com/home/index
8.2. Biomarker Discovery Using Systems Biology Approach.
NGSmakesitpossibletogeneratemultipletypesofgenomicalterations, including mutations, gene fusions, copy num-ber alterations, and epigenetic changes simultaneously ina single test. Integration of these genomic, transcriptomic,interactomic,andepigenomicpiecesofinformationisessen-tial to infer the underlying mechanisms in prostate cancerdevelopment. The challenge ahead will be developing acomprehensiveapproachthatcouldbeanalysedacrossthesecomplementary data, looking for an ideal combination ofbiomarkersignatures.
Consequently,thefutureofbiomarkerdiscoverywillrely
on a systems biology approach. One of the most fascinatingfields in this regard is the network-based approaches tobiomarker discovery, which integrate a large and heteroge-neous dataset into interactive networks. In such networks,molecular components and interactions between them arerepresentedasnodesandedges,respectively.Theknowledgeextracted from different types of networks can assist discov-ery of novel biomarkers, for example, functional pathways,processes,orsubnetworks,forimproveddiagnostic,prognos-tic, and drug response prediction. Network-based discoveryframework has already been reported in several types of
cancers [98–102]i n c l u d i n gP C a[ 103,104]. In a pioneering
study, Jin et al. [ 103] built up a prostate-cancer-related net-
work(PCRN)bysearchingtheinteractionsamongidentifiedmoleculesrelatedtoprostatecancer.Thenetworkbiomarkersderivedfromthenetworkdisplayhigh-performancesinPCapatientclassification.
As the field high-throughput technologies continues to
develop,wewillexpectenhancingcooperationamongdiffer-entdisciplinesintranslationalbioinformatics,suchasbioin-formatics,imaginginformatics,clinicalinformatics,andpub-lichealthinformatics.Notably,theheterogeneousdatatypescoming from various informatics platforms are pushing fordeveloping standards for data exchange across specialized
domains.
8.3. Personalized Biomarkers. The recent breakthroughs in
NGS also promise to facilitate the area of “personalised”biomarkers. Prostate cancer is highly heterogeneous amongindividuals. Current evidence indicates that the inter-individualheterogeneityarisesfromgeneticalenvironmentaland lifestyle factors. By deciphering the genetic make-upof prostate tumors, NGS may facilitate patient stratificationfor targeted therapies and therefore assist tailoring the besttreatmenttotherightpatient.Itisenvisionedthatpersonal-izedtherapywillbecomepartofclinicalpracticeforprostatecancerinthenearfuture.
9. Conclusions
Technological advances in NGS have increased our knowl-edge in molecular basis of PCa. However the translation ofmultiple molecular markers into the clinical realm is in itsearly stages. The full application of translational bioinfor-maticsinPCadiagnosisandprognosisrequirescollaborativeefforts between multiple disciplines. We can envision that
thecloud-supportedtranslationalbioinformaticsendeavours
will promote faster breakthroughs in the diagnosis andprognosisofprostatecancer.
Conflict of Interests
Theauthorsdeclarethatthereisnoconflictofinterests.
Authors’ Contribution
Jiajia Chen, Daqing Zhang, and Wenying Yan contributedequallytothiswork.

10 BioMedResearchInternational
Acknowledgments
The authors gratefully acknowledge financial support from
the National Natural Science Foundation of China Grants(91230117, 31170795), the Specialized Research Fund forthe Doctoral Program of Higher Education of China(20113201110015),InternationalS&TCooperationProgramofSuzhou(SH201120),theNationalHighTechnologyResearchand Development Program of China (863 program, Grantno. 2012AA02A601), and Natural Science Foundation forCollegesandUniversitiesinJiangsuProvince(13KJB180021).
References
[ 1 ]S .C a r l s s o n ,A .J .V i c k e r s ,M .R o o b o le ta l . ,“ P r o s t a t ec a n c e r
screening: facts, statistics, and interpretation in response to
theuspreventiveservicestaskforcereview,” J o u rn a lo fCl i n i c a l
Oncology,vol.30,no .21,pp .2581 –2584,2012.
[2] L.D.F .V enderbosandM.J.Roobol,“PSA-basedprostatecancer
screening: the role of active surveillance and informed andshareddecisionmaking,” AsianJournalofAndrology ,vol.13,no.
2,pp.219–224,2011.
[3] O.Sartor,“Randomized studiesofPSAscreening: anopinion,”
AsianJournalofAndrology ,vol.13,no .3,pp .364–365,2011.
[4] I. N. Sarkar, “Biomedical informatics and translational
medicine,” Journal of Translational Medicine ,v o l .8 ,a r t i c l e2 2 ,
2010.
[5] C.A.KulikowskiandC.W .Kulikowski,“Biomedicalandhealth
informatics in translational medicine,” Methods of Information
inMedicine ,vol.48,no .1,pp .4–10,2009 .
[6] B. S. Taylor, N. Schultz, H. Hieronymus et al., “Integrative
genomic profiling of human prostate cancer,” Cancer Cell ,v o l .
18,no .1,pp .11 –22,2010.
[7] W.Liu,S.Laitinen,S.Khanetal.,“Copynumberanalysisindi-
cates monoclonal origin of lethal metastatic prostate cancer,”
Nature Medicine ,vol.15,no .5,pp .559–565,2009 .
[8] X. Mao, Y. Yu, L. K. Boyd et al., “Distinct genomic alterations
in prostate cancers in Chinese and Western populations sug-
gest alternative pathways of prostate carcinogenesis,” Cancer
Research,vol.70,no .13,pp .5207 –5212,2010.
[ 9 ]T .W .F r i e d l a n d e r ,R .R o y ,S .A .T o m l i n se ta l . ,“ C o m m o n
structural and epigenetic changes in the genome of castration-resistant prostate cancer,” Cancer Research ,v o l .7 2 ,n o .3 ,p p .
616–625,2012.
[10] J.Sun,W.Liu,T.S.Adamsetal.,“DNAcopynumberalterations
in prostate cancers: a combined analysis of published CGH
studies,”Prostate,vol.67 ,no .7 ,pp .692–700,2007 .
[11] C. M. Robbins, W. A. Tembe, A. Baker et al., “Copy number
and targeted mutational analysis reveals novel somatic eventsin metastatic prostate tumors,” Genome Research ,v o l .2 1 ,n o .1 ,
pp .47 –55,2011.
[12] M.J.Kim,R.D.Cardiff,N.Desaietal.,“CooperativityofNkx3.1
andPtenlossoffunctioninamousemodelofprostatecarcino-
genesis,”Proceedings of the National Academy of Sciences of the
UnitedStatesofAmerica ,vol.99 ,no .5,pp .2884–2889 ,2002.
[ 1 3 ]J .L u o ,S .Z h a ,W .R.G a g ee tal . ,“ 𝛼-methylacyl-CoA racemase:
a new molecular marker for prostate cancer,” Cancer Research ,
vol.62,no.8,pp.2220–2226,2002.
[14] M.Tinzl,M.Marberger,S.Horvath,andC.Chypre,“DD3PCA3
RNAanalysisinurine—anewperspectivefordetectingprostate
cancer,”EuropeanUrology ,vol.46,no .2,pp .182–187 ,2004.[15]E.S.Leman,G.W .Cannon,B.J.T rocketal.,“EPCA -2:ahighly
specificserummarkerforprostatecancer,” Urology,vol.69 ,no .
4,pp.714–720,2007.
[16] J. Luo, D. J. Duggan, Y. Chen et al., “Human prostate cancer
andbenignprostatichyperplasia:moleculardissectionbygene
expressionprofiling,” CancerResearch ,vol.61,no.12,pp.4683–
4688,2001.
[17] M. F. Darson, A. Pacelli, P. Roche et al., “Human glandular
kallikrein 2 (hK2) expression in prostatic intraepithelial neo-
plasia and adenocarcinoma: a novel prostate cancer marker,”Urology,vol.49 ,no .6,pp .857 –862,1997 .
[18] S. Varambally, S. M. Dhanasekaran, M. Zhou et al., “The
polycomb group protein EZH2 is involved in progression ofprostatecancer,” Nature,vol.419 ,no .6907 ,pp .624–629 ,2002.
[19] S. F. Shariat, B. Andrews, M.W. Kattan, J. Kim, T. M.Wheeler,
andK.M.Slawin,“Plasmalevelsofinterleukin-6anditssolublereceptor are associated with prostate cancer progression and
metastasis,” Urology,vol.58,no .6,pp .1008–1015,2001.
[20] A. Berruti, A. Mosca, M. Tucci et al., “Independent prognostic
role of circulating chromogranin A in prostate cancer patients
with hormone-refractory disease,” Endocrine-Related Cancer ,
vol.12,no .1,pp .109–117 ,2005.
[21] S. F. Shariat, C. G. Roehrborn, J. D. McConnell et al., “Asso-
ciation of the circulating levels of the urokinase system of
plasminogen activation with the presence of prostate cancer
and invasion, progression, and metastasis,” Journal of Clinical
Oncology,vol.25,no .4,pp .349–355,2007 .
[22] S. F. Shariat, M. W. Kattan, E. Traxel et al., “Association of pre-
and postoperative plasma levels of transforming growth factor𝛽1 and interleukin 6 and its soluble receptor with prostate
cancerprogression,” ClinicalCancerResearch ,vol.10,no .6,pp .
1992–1999,2004.
[23] S.F.Shariat,J.H.Kim,B.Andrewsetal.,“Preoperativeplasma
levels of transforming growth factor beta(1) strongly predict
clinical outcome in patients with bladder carcinoma,” Cancer,
vol.92,no .12,pp .2985–2992,2001.
[24] J.Lapointe,C.Li,J.P.Higginsetal.,“Geneexpressionprofiling
identifies clinically relevant subtypes of prostate cancer,” Pro-
ceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica ,vol.101,no .3,pp .811 –816,2004.
[25] G. Kristiansen, C. Pilarsky, C. Wissmann et al., “Expression
profiling of microdissected matched prostate cancer samplesreveals CD166/MEMD and CD24 as new prognostic markers
for patient survival,” Journal of Pathology ,v o l .2 0 5 ,n o .3 ,p p .
359–376,2005.
[26] J. Lapointe, S. Malhotra, J. P. Higgins et al., “hCAP-D3 expres-
sion marks a prostate cancer subtype with favorable clinical
behavior and androgen signaling signature,” American Journal
ofSurgicalPathology ,vol.32,no.2,pp.205–209,2008.
[ 2 7 ]K .D .S ø r e n s e n ,P .J .W i l d ,A .M o r t e z a v ie ta l . ,“ G e n e t i ca n d
epigeneticSLC18A2silencingin prostatecancerisanindepen-
dent adverse predictor of biochemical recurrence after radical
prostatectomy,” ClinicalCancerResearch ,vol.15,no.4,pp.1400–
1410,2009.
[28] J. F. Knight, C. J. Shepherd, S. Rizzo et al., “TEAD1 and c-Cbl
are novel prostate basal cell markers that correlate with poor
clinical outcome in prostate cancer,” British Journal of Cancer ,
vol.99 ,no .11,pp .1849–1858,2008.
[29] W. D. Zhong, G. Q. Qin, Q. S. Dai et al., “SOXs in human
prostate cancer: implication as progression and prognosis
factors,”BMCCancer ,vol.12,no .1,p .248,2012.

BioMedResearchInternational 11
[30] C.Ruiz,M.Oeggerli,M.Germannetal.,“HighNRBP1expres-
sioninprostatecancerislinkedwithpoorclinicaloutcomesand
increasedcancercellgrowth,” Prostate,2012.
[31] N.P´ertega-Gomes,J.R.Vizca ´ıno,V.Miranda-Gonc ¸alvesetal.,
“Monocarboxylate transporter 4 (MCT4) and CD147 overex-
pression is associated with poor prognosis in prostate cancer,”BMCCancer ,vol.11,p.312,2011.
[32] A. S. Syed Khaja, L. Helczynski, A. Edsj ¨oe ta l . ,“ E l e v a t e dl e v e l
ofWnt5aproteininlocalizedprostatecancertissueisassociatedwithbetteroutcome,” PloSONE ,vol.6,no.10,ArticleIDe26539,
2011.
[33] M.Q.Yang,B.D.Athey,H.R.Arabniaetal.,“High-throughput
next-generation sequencing technologies foster new cutting-
edgecomputingtechniquesinbioinformatics,” BMCGenomics ,
vol.10,no.1,articleI1,2009.
[34] C. C. Collins, S. V. Volik, A. V. Lapuk et al., “Next generation
sequencing of prostate cancer from a patient identifies a defi-
ciency of methylthioadenosine phosphorylase, an exploitabletumor target,” Molecular Cancer Therapeutics ,v o l .1 1 ,n o .3 ,p p .
775–783,2012.
[35] C.E.Barbieri,S.C.Baca,M.S.Lawrenceetal.,“Exomesequenc-
ing identifies recurrent SPOP, FOXA1 and MED12 mutations
inprostatecancer,” NatureGenetics ,vol.44,no .6,pp .685–689 ,
2012.
[36] C. S. Grasso, Y. M. Wu, D. R. Robinson et al., “The mutational
landscapeoflethalcastration-resistantprostatecancer,” Nature,
vol.487,no.7406,pp.239–243,2012.
[37] A.Kumar,T.A.White,A.P.MacKenzieetal.,“Exomesequenc-
ing identifies a spectrum of mutation frequencies in advancedand lethal prostate cancers,” Proceedings of the National
AcademyofSciencesoftheUnitedStatesofAmerica ,vol.108,no.
41,pp.17087–17092,2011.
[38] S.A.Tomlins,D.R.Rhodes,S.Perneretal.,“Recurrentfusion
of TMPRSS2 and ETS transcription factor genes in prostate
cancer,”Science,vol.310,no.5748,pp.644–648,2005.
[39] S.A.Tomlins,R.Mehra,D.R.Rhodesetal.,“TMPRSS2:ETV4
gene fusions define a third molecular subtype of prostate
cancer,”CancerResearch ,vol.66,no .7 ,pp .3396–3400,2006.
[40] B.E.Helgeson,S.A.Tomlins,N.Shahetal.,“Characterizationof
TMPRSS2:ETV5 and SLC45A3:ETV5 gene fusions in prostate
cancer,”CancerResearch ,vol.68,no .1,pp .73–80,2008.
[41] Z.Shaikhibrahim,M.Braun,P.Nikolovetal.,“Rearrangement
of the ETS genes ETV-1, ETV-4, ETV-5, and ELK-4 is a clonalevent during prostate cancer progression,” Human Pathology ,
vol.43,no .11,pp .1910–1916,2012.
[42] S. A. Tomlins, B. Laxman, S. M. Dhanasekaran et al., “Distinct
classes of chromosomal rearrangements create oncogenic ETS
gene fusions in prostate cancer,” Nature,v o l .4 4 8 ,n o .7 1 5 3 ,p p .
595–599,2007.
[43] K.G.Hermans,A.A.Bressers,H.A.VanDerKorput,N.F.Dits,
G.Jenster,andJ.Trapman,“Twouniquenovelprostate-specific
and androgen-regulated fusion partners of ETV4 in prostatecancer,”CancerResearch ,vol.68,no.9,pp.3094–3098,2008.
[44] N. Palanisamy, B. Ateeq, S. Kalyana-Sundaram et al., “Rear-
rangements of the RAF kinase pathway in prostate cancer,gastric cancer and melanoma,” Nature Medicine ,v o l .1 6 ,n o .7 ,
pp.793–798,2010.
[45]C.W u,A.W .W yatt,A.V .Lapuketal.,“Integratedgenomeand
transcriptomesequencingidentifiesanovelformofhybridand
aggressiveprostatecancer,” Journal of Pathology ,v o l .227 ,n o .1,
pp .53–61,2012.[46] H.Beltran,R.Yelensky,G.M.Framptonetal.,“Targetednext-
generation sequencing of advanced prostate cancer identifies
potential therapeutic targets and disease heterogeneity,” Euro-
peanUrology ,vol.63,no .5,pp .920–926,2013.
[47] K.Kannan,L.Wang,J.Wang,M.M.Ittmann,W.Li,andL.Yen,
“RecurrentchimericRNAsenrichedinhumanprostatecancer
identified by deep sequencing,”
Proceedings of the National
AcademyofSciencesoftheUnitedStatesofAmerica ,vol.108,no.
22,pp.9172–9177,2011.
[48] J. R. Prensner, M. K. Iyer, O. A. Balbin et al., “Transcriptome
sequencing across a prostate cancer cohort identifies PCAT-
1, an unannotated lincRNA implicated in disease progression,”Nature Biotechnology ,vol.29 ,no .8,pp .7 42–7 49 ,2011.
[49] A. Watahiki, Y. Wang, J. Morris et al., “MicroRNAs associated
withmetastaticprostatecancer,” PLoSONE ,vol.6,no.9,Article
IDe24950,2011.
[ 5 0 ]K .R .C h n g ,C .W .C h a n g ,S .K .T a ne ta l . ,“ At r a n s c r i p t i o n a l
repressor co-regulatory network governing androgen responseinprostatecancers,” EMBOJournal ,vol.31,pp.2810–2823,2012.
[51] Z.Zhang,C.W.Chang,W.L.Goh,W.-K.Sung,andE.Cheung,
“CENTDIST:discoveryofco-associatedfactorsbymotifdistri-
bution,”Nucleic Acids Research ,v o l.3 9 ,no .2,p p .W3 91 –W3 99 ,
2011.
[52] H.H.He,C.A.Meyer,M.W .Chen,V .C.Jordan,M.Brown,and
X. S. Liu, “Differential DNase I hypersensitivity reveals factor-
dependentchromatindynamics,” GenomeResearch ,vol.22,no.
6,pp.1015–1025,2012.
[ 5 3 ]G .H .L i t t l e ,H .N o u s h m e h r ,S .K .B a n i w a l ,B .P .B e r m a n ,G .
A. Coetzee, and B. Frenkel, “Genome-wide Runx2 occupancy
in prostate cancer cells suggests a role in regulating secretion,”
NucleicAcidsResearch ,vol.40,no .8,pp .3538–3547 ,2012.
[ 5 4 ]P .Y .T a n ,C .W .C h a n g ,K .R .C h n g ,K .D .S e n a l iA b a y r a t n a
Wansa, W.-K. Sung, and E. Cheung, “Integration of regulatory
networks by NKX3-1 promotes androgen-dependent prostate
cancer survival,” Molecular and Cellular Biology ,v o l .3 2 ,n o .2 ,
pp .399–414,2012.
[55] J. H. Kim, S. M. Dhanasekaran, J. R. Prensner et al., “Deep
sequencing reveals distinct patterns of DNA methylation in
prostatecancer,” GenomeResearch ,vol.21,no .7 ,pp .10 28–1041,
2011.
[56] B. Langmead, M. C. Schatz, J. Lin, M. Pop, and S. L. Salzberg,
“Searching for SNPs with cloud computing,” Genome Biology ,
vol.10,no.11,articleR134,2009.
[57] M.C.Schatz,“CloudBurst:highlysensitivereadmappingwith
MapReduce,” Bioinformatics ,vol.25,no.11,pp.1363–1369,2009.
[58] M. C. Schatz, D. D. Sommer, D. R. Kelley, and M. Pop, “De
Novo assembly of large genomes using cloud computing,” inProceedings of the Cold Spring Harbor Biology of Genomes
Conference ,N ewY ork,NY ,USA,May2010.
[ 5 9 ]A .K .T a l u k d e r ,S .G a n d h a m ,H .A .P r a h a l a d ,a n dN .P .
Bhattacharyya,“Cloud-MAQ:thecloud-enabledscalablewholegenome reference assembly application,” in Proceedings of the
7th IEEE and IFIP International Conference on Wireless and
Optical Communications Networks (WOCN ’10) ,C o l o m b o ,S r i
Lanka,September2010.
[60] M.H.Rahimi, Bioscope:A ctuatedSensorN etworkforBiological
Science, University of Southern California, Los Angeles, Calif,
USA,2005.
[61] C. Sansom, “Up in a cloud?” Nature Biotechnology ,v o l .2 8 ,n o .
1,pp.13–15,2010.

12 BioMedResearchInternational
[62] T.Nguyen,W.Shi,andD.Ruden,“CloudAligner:afastandfull-
featured MapReduce based tool for sequence mapping,” BMC
ResearchNotes ,vol.4,article171,2011.
[63] P. Kudtarkar, T. F. DeLuca, V. A. Fusaro, P. J. Tonellato, and
D. P. Wall, “Cost-effective cloud computing: a case study
using the comparative genomics tool, roundup,” Evolutionary
Bioinformatics ,vol.2010,no .6,pp .197 –203,2010.
[64] X. Feng, R. Grossman, and L. Stein, “PeakRanger: a cloud-
enabledpeakcallerforChIP-seqdata,” BMCBioinformatics ,vol.
12,article139,2011.
[65] B. Langmead, K. D. Hansen, and J. T. Leek, “Cloud-scale
RNA-sequencing differential expression analysis with Myrna,”GenomeBiology ,vol.11,no.8,ArticleIDR83,2010.
[66] A.Goncalves,A.Tikhonov,A.Brazma,andM.Kapushesky,“A
pipeline for RNA-seq data processing and quality assessment,”
Bioinformatics ,vol.27 ,no .6,pp .867 –869 ,2011.
[67] Y. Li and S. Zhong, “SeqMapReduce: software and web service
for accelerating sequence mapping,” Critical Assessment of
MassiveDataAnaysis(CAMDA) ,vol.2009 ,2009 .
[68] M.F.Berger,M.S.Lawrence,F.Demichelisetal.,“Thegenomic
complexityofprimaryhumanprostatecancer,” Nature,vol.470,
no .7333,pp .214–220,2011.
[69] I. N. Holcomb, D. I. Grove, M. Kinnunen et al., “Genomic
alterationsindicatetumororiginandvariedmetastaticpotential
of disseminated cells from prostate cancer patients,” Cancer
Research,vol.68,no.14,pp.5599–5608,2008.
[70] Z.Kan,B.S.Jaiswal,J.Stinsonetal.,“Diversesomaticmutation
patternsandpathwayalterationsinhumancancers,” Nature,vol.
466,no.7308,pp.869–873,2010.
[71] L. Agell, S. Hern ´andez, S. De Muga et al., “KLF6 and TP53
mutations are a rare event in prostate cancer: distinguishingbetweenTaqpolymeraseartifactsandtruemutations,” Modern
Pathology ,vol.21,no .12,pp .1470–1478,2008.
[72] J.-T. Dong, “Prevalent mutations in prostate cancer,” Journal of
CellularBiochemistry ,vol.97 ,no .3,pp .433–447 ,2006.
[73] X. Sun, H. F. Frierson, C. Chen et al., “Frequent somatic
mutationsofthetranscriptionfactorATBF1inhumanprostate
cancer,”NatureGenetics ,vol.37 ,no .6,pp .407 –412,2005.
[74] L. Zheng, F. Wang, C. Qian et al., “Unique substitution of
CHEK2 and TP53 mutations implicated in primary prostatetumors and cancer cell lines,” Human Mutation ,v o l .2 7 ,n o .1 0 ,
pp .1062–1063,2006.
[75] C.Kumar-Sinha,S.A.Tomlins,andA.M.Chinnaiyan,“Recur-
rent gene fusions in prostate cancer,” Nature Reviews Cancer ,
vol.8,no .7 ,pp .497 –511,2008.
[76] S.Perner,J.-M.Mosquera,F.Demichelisetal.,“TMPRSS2-ERG
fusionprostatecancer:anearlymoleculareventassociatedwith
invasion,” AmericanJournalofSurgicalPathology ,vol.31,no .6,
pp.882–888,2007.
[77] S. A. Tomlins, S. M. J. Aubin, J. Siddiqui et al., “Urine
TMPRSS2:ERG fusion transcript stratifies prostate cancer
risk in men with elevated serum PSA,” Science Translational
Medicine,vol.3,no .94,ArticleID94ra72,2011.
[78] D.B.Seligson,S.Horvath,T.Shietal.,“Globalhistonemodifica-
tionpatternspredictriskofprostatecancerrecurrence,” Nature,
vol.435,no.7046,pp.1262–1266,2005.
[79] A. Urbanucci, B. Sahu, J. Sepp ¨al¨a et al., “Overexpression of
androgen receptor enhances the binding of the receptor to the
chromatininprostatecancer, ” Oncogene ,vol.31,no.17 ,pp.2153–
2163,2012.[80] D.S.Rickman,T.D.Soong,B.Mossetal.,“Oncogene-mediated
alterations in chromatin conformation,” Proceedings of the
National Academy of Sciences of the United States of America
,
vol.109,no.23,pp.9083–9088,2012.
[81] X.Lin,M.Tascilar,W.-H.Leeetal.,“GSTP1CpGislandhyper-
methylationisresponsiblefortheabsenceofGSTP1expressioninhumanprostatecancercells,” A mericanJ ournalofPathology ,
vol.159,no.5,pp.1815–1826,2001.
[82] E.Rosenbaum,M.O.Hoque,Y.Cohenetal.,“Promoterhyper-
methylation as anindependent prognosticfactor for relapsein
patients with prostate cancer following radical prostatectomy,”
ClinicalCancerResearch ,vol.11,no .23,pp .8321 –8325,2005.
[83] O. Taiwo, G. A. Wilson, T. Morris et al., “Methylome analysis
usingMeDIP-seqwithlowDNAconcentrations,” NatureProto-
cols,vol.7 ,no .4,pp .617 –636,2012.
[84] R. Lister, M. Pelizzola, R. H. Dowen et al., “Human DNA
methylomes at base resolution show widespread epigenomic
differences,” Nature,vol.462,no .7271,pp .315–322,2009 .
[85] A.Meissner,A.Gnirke,G.W.Bell,B.Ramsahoye,E.S.Lander,
and R. Jaenisch, “Reduced representation bisulfite sequencingfor comparative high-resolution DNA methylation analysis,”NucleicAcidsResearch ,vol.33,no .18,pp .5868–5877 ,2005.
[86] C. Gebhard, L. Schwarzfischer, T.-H. Pham et al., “Genome-
wide profiling of CpG methylation identifies novel targets
of aberrant hypermethylation in myeloid leukemia,” Cancer
Research,vol.66,no .12,pp .6118–6128,2006.
[ 8 7 ]A .L .B r u n n e r ,D .S .J o h n s o n ,W .K .S ie ta l . ,“ D i s t i n c tD N A
methylationpatternscharacterizedifferentiatedhumanembry-
onic stem cells and developing human fetal liver,” Genome
Research,vol.19 ,no .6,pp .1044–1056,2009 .
[ 8 8 ]M .W .K a t t a n ,S .F .S h a r i a t ,B .A n d r e w se ta l . ,“ Th ea d d i t i o n
of interleukin-6 soluble receptor and transforming growthfactor beta1 improves a preoperative nomogram for predicting
biochemical progression in patients with clinically localized
prostatecancer,” JournalofClinicalOncology ,vol.21,no .19 ,pp .
3573–3579,2003.
[89] J. Baden, G. Green, J. Painter et al., “Multicenter evaluation of
aninvestigationalprostatecancermethylationassay,” Journalof
Urology,vol.182,no .3,pp .1186–1193,2009 .
[90] D.R.Rhodes,T.R.Barrette,M.A.Rubin,D.Ghosh,andA.M.
Chinnaiyan, “Meta-analysis of microarrays: interstudy valida-tion of gene expression profiles reveals pathway dysregulation
in prostate cancer,” Cancer Research ,v o l .6 2 ,n o .1 5 ,p p .4 4 2 7 –
4433,2002.
[91] G. V. Glinsky, A. B. Glinskii, A. J. Stephenson, R. M. Hoffman,
and W. L. Gerald, “Gene expression profiling predicts clinical
outcome of prostate cancer,” Journal of Clinical Investigation ,
vol.113,no.6,pp.913–923,2004.
[92] Y.Wang,J.Chen,Q.Lietal.,“Identifyingnovelprostatecancer
associatedpathwaysbasedonintegrativemicroarraydataanal-ysis,”Computational Biology and Chemistry ,v o l .3 5 ,n o .3 ,p p .
151–158,2011.
[93] L.-C.Li,H.Zhao,H.Shiina,C.J.Kane,andR.Dahiya,“PGDB:a
curatedandintegrateddatabaseofgenesrelatedtotheprostate,”
NucleicAcidsResearch ,vol.31,no .1,pp .291 –293,2003.
[94] V. Hawkins, D. Doll, R. Bumgarner et al., “PEDB: the prostate
expression database,” Nucleic Acids Research ,v o l .2 7 ,n o .1 ,p p .
204–208,1999.
[95] M. Maqungo, M. Kaur, S. K. Kwofie et al., “DDPC: dragon
databaseofgenesassociatedwithprostatecancer,” NucleicAcids
Research,vol.39 ,no .1,pp .D980–D985,2011.

BioMedResearchInternational 13
[96] A.Etim,G.Zhou,X.Wenetal.,“ChromSorterPC:adatabaseof
chromosomal regions associated with human prostate cancer,”
BMCGenomics ,vol.5,article27 ,2004.
[97] B.D.O’Connor,B.Merriman,andS.F.Nelson,“SeqWareQuery
Engine:storingandsearchingsequencedatainthecloud,” BMC
Bioinformatics ,vol.11,no .12,articleS2,2010.
[98] H.-Y.Chuang,E.Lee,Y.-T.Liu,D.Lee,andT.Ideker,“Network-
based classification of breast cancer metastasis,” Molecular
Systems Biology ,vol.3,article140,2007 .
[99] I.W.Taylor,R.Linding,D.Warde-Farleyetal.,“Dynamicmod-
ularity in protein interaction networks predicts breast cancer
outcome,” Nature Biotechnology ,v o l .2 7 ,n o .2 ,p p .1 9 9 – 2 0 4 ,
2009.
[100] M.Xu,M.-C.J.Kao,J.Nunez-Iglesias,J.R.Nevins,M.W est,and
X.J.Jasmine,“Anintegrativeapproachtocharacterizedisease-specificpathwaysandtheircoordination:acasestudyincancer,”
BMCGenomics ,vol.9 ,no .1,articleS12,2008.
[101] Y.-C. Wang and B.-S. Chen, “A network-based biomarker
approach for molecular investigation and diagnosis of lung
cancer,”BMCMedicalGenomics ,vol.4,article2,2011.
[ 1 0 2 ]Y .Z h a n g,S .W a n g,D .L ie ta l . ,“ As y s t e m sb i o l o g y – b a s e dc l a s –
sifier for hepatocellular carcinoma diagnosis,” PLoS ONE ,v o l .
6,no.7,ArticleIDe22426,2011.
[ 1 0 3 ]G .J i n ,X .Z h o u ,K .C u i ,X . – S .Z h a n g ,L .C h e n ,a n dS .T .C .
Wong, “Cross-platform method for identifying candidate net-
work biomarkers for prostate cancer,” IETSystemsBiology ,v o l .
3,no.6,pp.505–512,2009.
[104] R. Ummanni, F. Mundt, H. Pospisil et al., “Identification of
clinically relevant protein targets in prostate cancer with 2D-DIGEcoupledmassspectrometryandsystemsbiologynetwork
platform,” PLoSONE ,vol.6,no .2,ArticleIDe16833,2011.

Copyright
of
BioMed
Research
International
is
the
property
of
Hindawi
Publishing
Corporation
and
its
content
may
not
be
copied
or
emailed
to
multiple
sites
or
posted
to
a
listserv
without
the
copyright
holder's
express
written
permission.
However,
users
may
print,
download,
or
email
articles
for
individual
use.