26133.生物信息多序列比對(duì)算法研究

1、南京農(nóng)業(yè)大學(xué)碩士學(xué)位論文生物信息多序列比對(duì)算法研究姓名：蔣彥超申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：計(jì)算機(jī)應(yīng)用技術(shù)指導(dǎo)教師：葉錫君20100601THERESEARCHoFBIoLoGICALMUIJIPLESEQUENCEALIGNMENTALGORITHMABS。I’RAC’I’Sequencealignmentisthemostimportantmanipulationandthefundamentalinformationprocessing

2、methodinbioinformaticsAligninganumberofsequencesissignificantfordiscoveringfunctional，structural，andevolutionaryinformationinbiologicalsequencesPairwisealignmentalgorithmsareverysuccessfulatpresentandtheexistingmultiples

3、equencealignmentalgorithmsaremostlybasedonamathematicalmodelorbiologicalmodel，whichCannotguaranteetheresultsofalignmentisoptimal，butratheraapproximationTherefore，itisstillahotanddifficulttostudybothefficientandaccurateon

4、multiplesequencealignmentalgorithmAmultiplesequencealignmentbasedonparticleswarmoptimization(PSO)ofevolutionandthehiddenMarkovmodels(HMM)isproposedbecauseofthehighlyeffectiveperformanceoftheintellectualizedalgorithminpro

5、cessingthemultiplesequencealignmentThebasisofsequencealignmentandanalysisofthesequencealigmnentinthegappenaltyandeffectsofscoringmatrixthedefinitionofmultiplesequencealignmentandtheSPmodelwhichevaluatetheresultofthealign

6、mentaleintroducedThenclassicalClustalw————afamousprogressivemethodisanalyzedindetailasallexampleofprogressivealignmentalgorithmsOnthisbasis，theintelligentalgorithmofhiddenMarkovmodelandthethreebasicproblemsforHMMandtheso

7、lutionaredescribedThehiddenMarkovmodelsformultiplesequencealignmentisestablishedandthealgorithmicprocessisgiventosolvethemultiplesequencealignmentAfterthattheshortcomingsofthehiddenMarkovModeltosolvemultiplesequencealign

8、mentarestudiedAstheBaum—WelchalgorithmmayconvergetoalargeiterationstrideundesirablelocalminimumtheparticleswarmoptimizationalgorithmtooptimizethetrainingofthemodelisintroducedParticleSwarmOptimizationiSacollaborativegrou

9、pbasedrandomsearchalgorithmbysimulatedflockforagingbehaviorThentheshortcomingsofparticleswarmoptimizationinsolvingthetrainingofthehiddenMarkovmodelareconductedanda‘premature’phenomenoninparticleswarmoptimizationisdescrib