原创力文档- 1
- 0
- 约2.73千字
- 约 8页
- 2026-03-16 发布于北京
- 举报
3/13/20261Lecture#7:FASTALFASTABIOINF2051Fall2002
2DotPlotAlphachainvs.BetachainofHumanHemoglobin
3FASTAandLFASTAPearsonandLipman(1988)FASTA–programthatcalculatestheinitialandoptimalsimilarityscoresbetweentwosequencesLFASTA–programfordetectinglocalsimilarities–findsmultiplealignmentsbetweensmallerportionsoftwosequences
4TheFASTAalgorithmFoursteps:Identifyregionsofsimilarity:Usingthektupparameterwhichspecifies#consecutiveidentitiesrequiredinamatch10bestdiagonalregionsfoundbasedon#matchesanddistancebetweenmatchesRescoreregionsandidentifybestinitialregionsPAM250orotherscoringmatrixusedforrescoringthe10diagonalregionsidentifiedinstep1toallowforconservativereplacementsandrunsofidentitiesshorterthanktupForeachthebestdiagonalregions,identify“initialregion”thatisbestscoringsubregion
5TheFASTAalgorithmOptimallyjoininitialregionswithscoresTGiven:locationofinitialregions,scores,gappenaltyCalculateanoptimalalignmentofinitialregionsasacombinationofcompatibleregionswithmaximalscoreUseresultingscoretorankthelibrarysequencesSelectivitydegradationlimitedbyusinginitialregionsthatscoregreaterthansomethresholdTAlignthehighestscoringlibrarysequencesusingmodificationofglobalandlocalalignmentalgorithmsConsidersallpossiblealignmentsofthequeryandlibrarysequencethatfallswithinabandcenteredaroundthehighestscoringinitialregion
6LFASTAFASTA–reportsonlyonehighestscoringalignmentbetweentwosequencesLFASTA–localsequencecomparisontoolthatcanidentifymultiplelocalalignmentsbetween2sequencesOptimalalgorithmsforsensitivelocalsequencecomparisonarecomputationallyintensiveintermsoftimeandmemory
7LFASTAvs.FASTALFASTAusessamefirst2stepsforfindinginitialregionsasFASTA,except:Insteadofsaving10initialregions,LFASTAsavesalldiagonalregionswithsimil
文档评论（0）