简介：

简介：

简介：Traditionally,SQLquerylanguageisusedtosearchthedataindatabases.However,itisinappropriateforend-users,sinceitiscomplexandhardtolearn.Itistheneedofend-user,searchingindatabaseswithkeywords,likeinwebsearchengines.Thispaperpresentsasurveyofworkonkeywordsearchindatabases.ItalsoincludesabriefintroductiontotheSEEKERsystemwhichhasbeendeveloped.

简介：

简介：

简介：

简介：

简介：

简介：RMBexchangerateTheexchangerateofRenminbi,theChinesecurrency,witnessedanappreciationofaroundsixpercenttooneU.S.dollarthisyear.OnDecem-ber13,thecentralparityratestoodatRMB7.3568againstonedollar,accordingtotheChineseForeignExchangeTradingSystem,breakingthe7.36mark.

简介：

简介：~~

简介：

简介：

简介：~~

简介：

简介：

简介：

简介：AconitiLateralisRadixPraeparata(Fuzi)isacommonlyusedtraditionalChinesemedicineinclinicforitspotencyinrestoringyangandrescuingfromcollapse.Aconitialkaloids,mainlyincludingmonoester-diterpenoidaconitines(MDAs)anddiester-diterpenoidaconitines(DDAs),areconsideredtoactasbothbioactiveandtoxicconstituents.Inthepresentstudy,afeasible,economical,andaccurateHPLCmethodforsimultaneousdeterminationofsixalkaloidmarkersusingtheSingleStandardforDeterminationofMulti-Components(SSDMC)methodwasdevelopedandfullyvalidated.Benzoylmesaconinewasusedastheuniquereferencestandard.Thismethodwasprovenasaccurate(recoveryvaryingbetween97.5%-101.8%,RSD<3%),precise(RSD0.63%-2.05%),andlinear(R>0.9999)overtheconcentrationranges,andsubsequentlyappliedtoquantitativeevaluationof62batchesofsamples,amongwhich45batcheswerefromgoodmanufacturingpractice(GMP)facilitiesand17batchesfromthedrugmarket.Thecontentswerethenanalyzedbyprincipalcomponentanalysis(PCA)andhomogeneitytest.ThepresentstudyprovidedvaluableinformationforimprovingthequalitystandardofAconitiLateralisRadixPraeparata.ThedevelopedmethodalsohasthepotentialinanalysisofotherAconitumspecies,suchasAconitumcarmichaelii(preparedparentroot)andAconitumkusnezoffii(preparedroot).

简介：Althoughtheproteinsequence-structuregapcontinuestoenlargeduetothedevelopmentofhigh-throughputsequencingtools,theproteinstructureuniversetendstobecompletewithoutproteinswithnovelstructuralfoldsdepositedintheproteindatabank(PDB)recently.Inthiswork,weidentifyaproteinstructuraldictionary(Frag-K)composedofasetofbackbonefragmentsrangingfrom4to20residuesasthestructural"keywords"thatcaneffectivelydistinguishbetweenmajorproteinfolds.Wefirstlyapplyrandomizedspectralclusteringandrandomforestalgorithmstoconstructrepresentativeandsensitiveproteinfragmentlibrariesfromalargescaleofhigh-quality,non-homologousproteinstructuresavailableinPDB.Weanalyzetheimpactsofclusteringcut-offsontheperformanceofthefragmenthbraries.Then,theFrag-KfragmentsareemployedasstructuralfeaturestoclassifyproteinstructuresinmajorproteinfoldsdefinedbySCOP(StructuralClassificationofProteins).OurresultsshowthatastructuraldictionarywithN4004-to20-residueFrag-KfragmentsiscapableofclassifyingmajorSCOPfoldswithhighaccuracy.

简介：Inthispaperwepresentanextensionofthetopologyoptimizationmethodtoincludeuncertaintiesdur-ingthefabricationofmacro,microandnanostructures.Morespecifically,weconsiderdevicesthataremanufacturedusingprocesseswhichmayresultin(uniformly)toothin(eroded)ortoothick(dilated)structurescomparedtotheintendedtopology.ExamplesareMEMSdevicesmanufacturedusingetchingprocesses,nano-devicesmanufacturedusinge-beamlithographyorlasermicro-machiningandmacrostructuresmanufacturedusingmillingprocesses.Inthesuggestedrobusttopologyoptimizationapproach,under-andover-etchingismodelledbyimageprocessing-based'erode'and'dilate'operatorsandtheoptimizationproblemisfor-mulatedasaworstcasedesignproblem.Applicationsofthemethodtothedesignofmacrostructuresforminimumcomplianceandmicrocompliantmechanismsshowthatthemethodprovidesmanufacturingtolerantdesignswithlittledecreaseinperformance.Asapositivesideeffecttherobustdesignformulationalsoeliminatesthelongstandingprob-lemofone-nodeconnectedhingesincompliantmechanismdesignusingtopologyoptimization.