资源描述：
《融合位置特征与序列进化信息的磷酸化位点预测（英文）》由会员上传分享，免费在线阅读，更多相关内容在工程资料-天天文库。
1、融合位置特征与序列进化信息的磷酸化位点预测(英文)谭泗桥李钎陈渊彭剑湖南农业大学信息科学技术学院湖南省农村农业信息化工程技术研究中心湖南农业大学信息化建设与管理中心湖南农业大学植物保护学院摘要：磷酸化是蛋白质翻译后的主要修饰，可分为激酶特异性和非激酶特异性两种类型.以非激酶特异性磷酸化位点Dou数据集为基础，本文发展了一种基于位置的卡方差表特征x2-pos,融合伪氨基酸序列进化信息PscPSSM表征序列，构建正负样本均衡的支持向量机分类器，S,T,Y独立测试Matthew相关系数、ROC曲线下面积分及准确率分别达到了(0.59、0
2、.87、79.74%),(0.55、0.85、77.68%)和(0.50、0.81、75.22%),明显优于文献报道结果.x2-pos>PsePSSM两种特征的融合在蛋口质磷酸化位点预测屮有广泛应用前景.关键词：磷酸化;预测；卡方差表特征;伪氨基酸序列进化信息;支持向量机;作者简介:谭泗桥Tel:013974893577,E-mail:545534721@qq.com收稿日期:July9,2017基金：supportedbyagrantfromNaturalScicnccFoundationofHunanprovince(14JJ
3、2082)PhosphorylationSitePredictionIntegratingThePositionFeatureWithSequenceEvolutionInformationTANSi-QiaoLIQianCHENYuanPENGTianCollegeofInformstionScienceandTechnology,HunanAgriculturalUniversity;HunanEngineeringResearchCenterforInformationTechnologyinAgricultureandRu
4、ral;CollegeofPlantProtection,HunanAgriculturalUniversity;Abstract：Phosphorylationisthemajorpost-translationmodificationtoproteins,anditcambeclassifiedaskinase-specificandnon-kinase-specific.Thispaperfocusesonthepredictionmethodsofnon-kinase-specificityandusingDou"sdat
5、asetofphosphorylationsitesasthetemplate,thispaperdevelopsaposition-basedchi-squaretablefeature,x2-pos,andthenintegratesthisfeaturewiththepseudoposition-specificscoringmatrix(PsePSSM)•ASupportVectorMachine(SVM)classifierwithbalancedpositiveandnegativesampleswascreated,
6、andtheS,T,YindependenttestingresultsfortheMatthewcorrelationcoefficient,theinferiorsurfaceintegraloftheROCcurveandtheprecisionwere(0.59,0.87,79.74%),(0.55,0.85,77.68%)and(0.50,0.81,75.22%),respectively,whicharcsignificantlysuperiortotheresuItsreportedpreviously.Theint
7、egrationofthex2-posandthePsePSSMoffersapromisingmethodtopredictphosphorylationsitesmoreaccuratelyinproteins.Keyword：phosphorylation;prediction;chi-squaretablefeature;PsePSSM;SVM;Received：July9,2017ProteinphosphorylationinvolvesthetransferofphosphategroupsofATPorGTPtos
8、pecificsitesonproteins,namelyserine,threonineandtyrosineresidues,byproteinkinases.Theprocesstransformsexternalstimuliintoint
