简介：Protein-proteininteractions(PPIs)havebeenwidelystudiedtounderstandthebiologicalprocessesormolecularfunctionsassociatedwithdifferentdiseasesystemslikecancer.Whilefocusedstudiesonindividualcancershavegeneratedvaluableinformation,globalandcomparativeanalysisofdatasetsfromdifferentcancertypeshasnotbeendone.Inthiswork,wecarriedoutbioinformaticanalysisofPPIscorrespondingtodifferentiallyexpressedgenesfrommicroarraysofvarioustumortissues(belongingtobladder,colon,kidneyandthyroidcancers)andcomparedtheirassociatedbiologicalprocessesandmolecularfunctions(basedonGeneOntologyterms).Weidentifiedasetofprocessesorfunctionsthatarecommontoallthesecancers,aswellasthosethatarespecifictoonlyoneorpartialcancertypes.Similarly,proteininteractionnetworksinnucleicacidmetabolismwerecomparedtoidentifythecommon/specificclustersofproteinsacrossdifferentcancertypes.Ourresultsprovideabasisforfurtherexperimentalinvestigationstostudyproteininteractionnetworksassociatedwithcancer.Themethodologydevelopedinthisworkcanalsobeappliedtostudysimilardiseasesystems.

简介：Arraycomparativegenomichybridization(CGH)hasbeenpopularlyusedforanalyzingDNAcopynumbervariationsindiseaseslikecancer.Inthisstudy,weinvestigated82sporadicsamplesfrom49breastcancerpatientsusing1-MbresolutionbacterialartificialchromosomeCGHarrays.Anumberofhighlyfrequentgenomicaberrationswerediscovered,whichmayactas'drivers'oftumorprogression.Meanwhile,thegenomicprofilesoffour'normal'breasttissuesamplestakenatleast2cmawayfromtheprimarytumorsiteswerealsofoundtohavesomegenomicaberrationsthatrecurredwithhighfrequencyintheprimarytumors,whichmayhaveimportantimplicationsforclinicaltherapy.Additionally,weperformedclasscomparisonandclasspredictionforvariousclinicopathologicalparameters,andalistofcharacteristicgenomicaberrationsassociatedwithdifferentclinicopathologicalphenotypeswascompiled.Ourstudyprovidescluesforfurtherinvestigationsoftheunderlyingmechanismsofbreastcarcinogenesis.

简介：二维的polyacrylamide胶化电气泳动(2D页)并且帮助矩阵的激光解吸附作用/电离双人脚踏车time-of-flight团spectrometry(MALDI-TOF/TOF-MS)，与联机数据库寻找合并了，被执行调查乳癌和邻近的正常的微分蛋白质胸纸巾。考虑到那浆液白朊富有地在正常被介绍，控制样品，从12在11检测的15个微分点(91.7%)乳癌样品被联机SIENA-2DPAGE数据库寻找和MALDI-TOF/TOF-MS分析识别。结果显示乳癌的病理学的变化涉及物质新陈代谢的扩大，解朊的活动的提升，酶的一些禁止者的活动的衰落等等。与改变的表示涉及乳癌的病理学的进程的一些重要蛋白质可以是有用简历标记，例如alpha-1-antitrypsin，EF-1-beta，组织蛋白酶D，TCTP，SMT3A，RPS12，和PSMA1，SMT3A，RPS12，和PSMA1首先在这研究为乳癌在之中被报导。

简介：Inthisstudy,wepresentaconstructivealgorithmfortrainingcooperativesupportvectormachineensembles(CSVMEs).CSVMEcombinesensemblearchitecturedesignwithcooperativetrainingforindividualSVMsinensembles.Unlikemostpreviousstudiesontrainingensembles,CSVMEputsemphasisonbothaccuracyandcollaborationamongindividualSVMsinanensemble.AgroupofSVMsselectedonthebasisofrecursiveclassifiereliminationisusedinCSVME,andthenumberoftheindividualSVMsselectedtoconstructCSVMEisdeterminedby10-foldcross-validation.ThiskindofSVMEhasbeentestedontwoovariancancerdatasetspreviouslyobtainedbyproteomicmassspectrometry.BycombiningseveralindividualSVMs,theproposedmethodachievesbetterperformancethantheSVMEofallbaseSVMs.

简介：Computationalanalysisisessentialfortransformingthemassesofmicroarraydataintoamechanisticunderstandingofcancer.Herewepresentamethodforfindinggenefunctionalmodulesofcancerfrommicroarraydataandhaveappliedittocoloncancer.First,acoloncancergenenetworkandanormalcolontissuegenenetworkwereconstructedusingcorrelationsbetweenthegenes.Thenthemodulesthattendedtohaveahomogeneousfunctionalcompositionwereidentifiedbysplit-tingupthenetwork.Analysisofbothnetworksrevealedthattheyarescale-free.Comparisonofthegenefunctionalmodulesforcoloncancerandnormaltissuesshowedthatthemodules’functionschangedwiththeirstructures.

简介：Quantitativegeneexpressionanalysisplaysanimportantroleinidentifyingdifferentiallyexpressedgenesinvariouspathologicalstates,geneexpressionregulationandco-regulation,sheddinglightongenefunctions.Althoughmicroarrayiswidelyusedasapowerfultoolinthisregard,itissuboptimalquantitativelyandunabletodetectunknowngenevariants.Herewedemonstratedeffectivedetectionofdifferentialexpressionandco-regulationofcertaingenesbyexpressedsequencetaganalysisusingaselectedsubsetofcDNAlibraries.Wediscussedtheissuesofsequencingdepthandlibrarypreparation,andproposethatincreasedsequencingdepthandimprovedpreparationproceduresmayallowdetectionofmanyexpressionfeaturesforlessabundantgenevariants.Withthereductionofsequencingcostandtheemergingofnewgenerationsequencingtechnology,in-depthsequencingofcDNApoolsorlibrariesmayrepresentabetterandpowerfultoolingeneexpressionprofilingandcancerbiomarkerdetection.Wealsoproposeusingsequence-specificsubtractiontoremovehundredsofthemostabundanthousekeepinggenestoincreasesequencingdepthwithoutaffectingrelativeexpressionratioofothergenes,astranscriptsfromasfewas300mostabundantlyexpressedgenesconstituteabout20%ofthetotaltranscriptome.In-depthsequencingalsorepresentsauniqueadvantageofdetectingunknownformsoftranscripts,suchasalternativesplicingvariants,fusiongenes,andregulatoryRNAs,aswellasdetectingmutationsandpolymorphismsthatmayplayimportantrolesindiseasepathogenesis.