資源描述：
《基于帶變異算子粒子群的多序列比對算法研究與應(yīng)用》由會員上傳分享，免費在線閱讀，更多相關(guān)內(nèi)容在學(xué)術(shù)論文-天天文庫。
1、摘要序列比對是生物信息學(xué)中一項重要的基礎(chǔ)性研究課題，多序列比對則是其最基本任務(wù)之一。由于各種原因，目前還沒有一個最佳的多序列比對算法。在研究比較了現(xiàn)有的多序列比對算法的基礎(chǔ)上，本文首次提出了使用帶變異算子粒子群優(yōu)化算法來解決多序列比對問題。本文首先分析了空位罰分、替換矩陣和目標(biāo)函數(shù)對序列比對的影響，具體實現(xiàn)SP和COFFEE目標(biāo)函數(shù)。然后通過對多序列比對算法的現(xiàn)狀的研究以及對粒子群優(yōu)化算法特點等的分析，實現(xiàn)了基于粒子群優(yōu)化算法的多序列比對算法MSAPSO(aParticleSwarmOptimizationdedicatedforMultipleSequenceAlign
2、ments)，提出了基于帶變異算子粒子群優(yōu)化算法的多序列比對算法MSAVPSO(aVariation-OperatorParticleSwarmOptimizationdedieatedforMultipleSequenceAlignments)，并分別實現(xiàn)了基于SP和COFFEE目標(biāo)函數(shù)的MSAPSO和MSAVPSO。最后用基準(zhǔn)多序列比對庫BALIBASEl．0中的用例對算法進(jìn)行測試，結(jié)果表明MSAVPSO算法在解決基因序列比對問題上是有效的。關(guān)鍵詞：：生物信息學(xué)多序列比對目標(biāo)函數(shù)粒子群優(yōu)化算法變異算子AbstraetAbstractSequencealignmento
3、fbioinformaticsisanimportantfundamentalsubjectinbioinformaticsresearch，oneofitsmostbasictaskistomultiplesequencealignments．Alsothereisnotanoptimalalgorithmofmultiplesequencealignments．Thispaperproposesamethod,whichUseParticleSwarmOptimizationandVariation-OperatorParticleSwarmOptimizationa
4、lgorithmofevolutionaryalgorithm，tosolvetheproblemsofmultiplesequencealignments．First,theeffectonsequencealignmentcausedbythegappenalty,substitutionmatrixandobjectivefunctionareanalyzed，andtheSPobjectivefunctionandCOFFEEobjectivefunctionareimplemented．Thenthedynamicprogrammingalgorithmthat
5、isoneofaccuratealgorithmsofpairwisealignmentisstudiedandimplemented．Laterthroughthestudyoncurrentsituationofthemultiplesequencealignmentsalgorithmandtheanalysistotheprincir'leandcharacteristicofthegeneticalgorithmandparticleswarmoptimizationalgorithmandSOon,presentthemultiplesequencealign
6、mentsalgorithmbasedonparticleswarmoptimizationalgorithmMSAPSO(aParticleSwarmOptimizationdedicatedforMultipleSequenceAlignment)andthemultiplesequencealignmentsalgorithmbasedonVariation-OperatorparticleswarmoptimizationalgorithmMSAVPSO(aVariation-OperatorParticleSwarmOptimizationdedicatedfo
7、rMultipleSequenceAlignment)．AlsothetwoalgorithmsbasedonSPobjectivefunctionandCOFFEEobjectivefunctionisimplemented．Finally,twoalgorithmsareusedtotestbenchmarkmultiplesequencealignmentsdatabaseBALIBASEl．0．TheresultsshowthattheproposedMSVAPSOalgorithmisfeasibletosolvet
