药物分子设计第九讲.ppt

《药物分子设计第九讲.ppt》由会员上传分享，免费在线阅读，更多相关内容在PPT专区-天天文库。

1、蛋白质结构预测(一)ProteinStructurePrediction(I)1ProteinStructurePredictionFromSequencetoNativeStructure2ProteinStructurePredictionPSI-BLAST3ProteinStructurePrediction4MultipleSequenceAlignmentToolsClustalWMultiplesequencealignmentsreveal:conservationofindividualresiduesconservationofregions

2、differenceswithinproteinfamilies5ToolsHMMerProfileForeachpositionalongthesequence,tabulatehowofteneachtypeofaminoacidoccur(include‘.’forgap)TheprofileisalwaysofsizeNx21,nomatterhowmanysequencesareconsidered6ProteinStructurePrediction7SecondaryStructurePredictionGivenaproteinsequence

3、a1a2…aN,secondarystructurepredictionaimsatdefiningthestateofeachaminoacidaiasbeingeitherH(helix),E(extendedstrand),orO(other)(Somemethodshave4states:H,E,Tforturns,andOforother).ThequalityofsecondarystructurepredictionismeasuredwithaQindex(Qhelix,Qstrand,Qcoil)orQ3.Q3isthepercentofre

4、siduesthatmatch“reality”(X-raystructure).8SecondaryStructurePredictionAminoacidsequenceActualSecondaryStructureQ3=22/29=76%Q3=22/29=76%ALHEASGPSVILFGSDVTVPPASNAEQAKhhhhhooooeeeeoooeeeooooohhhhhohhhooooeeeeoooooeeeooohhhhhhhhhhhoooohhhhooohhhooooohhhhh(usefulprediction)(terriblepredi

5、ction)Q3forrandompredictionis33%Secondarystructureassignmentinrealproteinsisuncertaintoabout10%.Therefore,a“perfect”predictionwouldhaveQ3=90%.LimitationsofQ3(Q3=Npredicted100%/Nobserved)9SecondaryStructurePredictionChou-FasmanMethodStartbycomputingaminoacidspropensitiestobelongtoag

6、iventypeofsecondarystructure:Propensities>1meanthattheresiduetypeiislikelytobefoundinthecorrespondingsecondarystructuretype.10Chou-FasmanMethod参数值P、P和Pt是分别在原有相应值的基础上乘以100得到SecondaryStructurePrediction11SecondaryStructurePredictionChou-FasmanMethodHelixprediction沿着蛋白质序列寻找螺旋核，相邻的6个

7、残基中如果有至少4个残基倾向于形成螺旋，即有4个残基对应的P>100，则认为是螺旋核。然后从螺旋核向两端延伸，直至四肽片段P的平均值小于100为止。按上述方式找到的片段长度大于5，并且P的平均值大于P的平均值，那么这个片段的二级结构就被预测为螺旋。此外，不容许Pro在螺旋内部出现，但可出现在C末端以及N端的前三位，这也用于终止螺旋的延伸。12SecondaryStructurePredictionChou-FasmanMethodStrandPrediction如果相邻6个残基中若有4个倾向于形成β折叠，即有4个残基对应的P>100，则认为是折叠

8、核。折叠核向两端延伸直至4个残基P的