""" CCP4PerformanceData.py: CCP4 GUI Project Copyright (C) 2014 STFC This library is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License version 3, modified in accordance with the provisions of the license to address the requirements of UK law. You should have received a copy of the modified GNU Lesser General Public License along with this library. If not, copi es may be downloaded from http://www.ccp4.ac.uk/ccp4license.php This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details. """ """ Liz Potterton Mar 2014 - performance indicators """ import CCP4Data,CCP4XtalData from PyQt4 import QtCore from CCP4ErrorHandling import * def performanceIndicatorClasses(): # This is used by CCP4DbApi.getJobPerformance (and perhaps other methods) to get the # appropriate classes from XData table return "('CPerformanceIndicator','CRefinementPerformance','CModelBuildPerformance','CDataReductionPerformance','CTestObsConversionsPerformance','CExpPhasPerformance','CPhaseErrorPerformance','CAtomCountPerformance')" # *************** new performance classes also need the keytype to be registered with the database ************************** # See the definition of KEYTYPELIST in dbapi/CCP4DbApi.py class CPerformanceIndicator(CCP4Data.CData): # Performance indicator expected in wrapper/pipeline output # This class should be reimplemented if value is not a float CONTENTS_ORDER = ['value','annotation'] CONTENTS = { 'value' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'annotation' : { 'class' : CCP4Data.CString } } ERROR_CODES = { 300 : { 'description' : 'Passed', 'severity': SEVERITY_OK }, 301 : { 'description' : 'Data value not set' }, 302 : { 'description' : 'Performance indicator value difference greater than tolereance' }, 303 : { 'description' : 'Performance indicator value different' }, 304 : { 'description' : 'Performance indicator value difference greater than tolereance - but improved', 'severity' : SEVERITY_WARNING } } def saveToDb(self): # Assume performance indicator is saved to db irespective of the saveToDb qualifier return [],None,{} # To support the QTreeItem.data() calls from CCP4ProjectWidget.CTreeItemJob def data(self,mode=QtCore.Qt.DisplayRole): if mode == QtCore.Qt.DisplayRole: return QtCore.QVariant(self.__str__()) return QtCore.QVariant() def scale(self,valueList,absoluteScale=True): # Not sure what this does yet. Probably need to scale a list of PIs of the same type # to determine colouring or whatever graphical presentation in the gui # Scaling could be absolute of per project pass def assertSame(self,other,testItems=[],diagnostic=False): # Return CErrorReport list any failed/warning differences intended for automatic testing # other should be another CPerformanceIndicator object # It is assumed that self is the original source project err = CErrorReport() # better is integer: 0=null, 1= increasing PI is improvement, -1= decreasing PI is improvement for item,tolerance,better in testItems: name = self.objectName()+'.'+item if diagnostic: print 'Comparing ',item,self.__dict__['_value'][item].objectPath(),'with tolerance',tolerance if not self.__dict__['_value'][item].isSet(): err.append(CPerformanceIndicator,301,name=name,details='in original project',stack=False) elif not other.__dict__['_value'][item].isSet(): err.append(CPerformanceIndicator,301,name=name,details='in test project',stack=False) else: if isinstance(self.__dict__['_value'][item],CCP4Data.CFloat): diff = float(other.__dict__['_value'][item])-float(self.__dict__['_value'][item]) if abs(diff)>tolerance: if better!=0 and better*diff>0.0: err.append(CPerformanceIndicator,304,name=name,details= str(self.__dict__['_value'][item])+' : '+str(other.__dict__['_value'][item]),stack=False) else: err.append(CPerformanceIndicator,302,name=name,details= str(self.__dict__['_value'][item])+' : '+str(other.__dict__['_value'][item]),stack=False) elif isinstance(self.__dict__['_value'][item],CCP4Data.CInt): diff = int(other.__dict__['_value'][item])-int(self.__dict__['_value'][item]) if abs(diff)>tolerance: if better!=0 and better*diff>0.0: err.append(CPerformanceIndicator,304,name=name,details= str(self.__dict__['_value'][item])+' : '+str(other.__dict__['_value'][item]),stack=False) else: err.append(CPerformanceIndicator,302,name=name,details= str(self.__dict__['_value'][item])+' : '+str(other.__dict__['_value'][item]),stack=False) elif isinstance(self.__dict__['_value'][item],CCP4Data.CString): if self.__dict__['_value'][item] != other.__dict__['_value'][item]: err.append(CPerformanceIndicator,303,name=name,details= str(self.__dict__['_value'][item])+' : '+str(other.__dict__['_value'][item]),stack=False) if len(err) == 0: err.append(CPerformanceIndicator,300,name=name) return err class CRefinementPerformance(CPerformanceIndicator): CONTENTS_ORDER = ['RFactor','RFree','RMSBond','RMSAngle','weightUsed','annotation'] CONTENTS = { 'RFactor' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'RFree' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'RMSBond' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'RMSAngle' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'weightUsed' : { 'class' : CCP4Data.CFloat }, 'annotation' : { 'class' : CCP4Data.CString } } def __str__(self): text = '' if self.__dict__['_value']['RFactor'].__dict__['_value'] is not None: text = text + 'R=' + format(self.__dict__['_value']['RFactor'],'.2f')+' ' if self.__dict__['_value']['RFree'].__dict__['_value'] is not None: text = text + 'RFree=' + format(self.__dict__['_value']['RFree'],'.2f')+' ' return text def saveToDb(self): #Return a list of files, string representation of xml, dict of key,value pairs #The dict keys must map to those defined in CCP4DbApi. ret = {} if self.RFactor.isSet(): ret['RFactor'] = self.RFactor.__float__() if self.RFree.isSet(): ret['RFree'] = self.RFree.__float__() return [],None,ret def assertSame(self,other,diagnostic=False): return CPerformanceIndicator.assertSame(self,other,[['RFactor',0.01,-1]],diagnostic=diagnostic) class CModelBuildPerformance(CPerformanceIndicator): CONTENTS_ORDER = ['RFactor','completeness','annotation'] CONTENTS = { 'RFactor' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'completeness' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'annotation' : { 'class' : CCP4Data.CString } } def __str__(self): text = '' if self.__dict__['_value']['RFactor'].__dict__['_value'] is not None: text = text + 'R=' + format(self.__dict__['_value']['RFactor'],'0.2f')+' ' if self.__dict__['_value']['completeness'].__dict__['_value'] is not None: text = text + ' %=' + format(int(self.__dict__['_value']['completeness']*100),'d') return text def saveToDb(self): #Return a list of files, string reporesentation of xml, dict of key,value pairs #The dict keys must map to does defined in CCP4DbApi. #print 'CModelBuildPerformance.saveToDb',{'RFactor' : self.RFactor.__float__(), 'completeness' : self.completeness.__float__() } ret = {} if self.RFactor.isSet(): ret['RFactor'] = self.RFactor.__float__() if self.completeness.isSet(): ret['completeness'] = self.completeness.__float__() return [],None,ret def assertSame(self,other,diagnostic=False): return CPerformanceIndicator.assertSame(self,other,[['RFactor',0.01,-1],['completeness',0.01,1]],diagnostic=diagnostic) class CDataReductionPerformance(CPerformanceIndicator): CONTENTS_ORDER = [ 'spaceGroup', 'highResLimit', 'rMeas' ] CONTENTS = { 'spaceGroup' : { 'class' : CCP4XtalData.CSpaceGroup }, 'highResLimit' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'rMeas' : { 'class' : CCP4Data.CFloat } } def __str__(self): text = '' if self.__dict__['_value']['spaceGroup'].isSet(): text += 'Sgp='+str(self.__dict__['_value']['spaceGroup']) if self.__dict__['_value']['highResLimit'].isSet(): text += ' res=' + format(self.__dict__['_value']['highResLimit'],'.2f') if self.__dict__['_value']['rMeas'].isSet(): text += ' Rmeas=' + format(self.__dict__['_value']['rMeas'],'.3f') return text def saveToDb(self): ret = { } if self.__dict__['_value']['spaceGroup'].isSet(): ret['spaceGroup'] = str(self.__dict__['_value']['spaceGroup']) if self.__dict__['_value']['highResLimit'].isSet(): ret['highResLimit'] = self.highResLimit.__float__() if self.__dict__['_value']['rMeas'].isSet(): ret['rMeas'] = self.rMeas.__float__() #print 'CDataReductionPerformance.saveToDb',ret return [],None,ret def assertSame(self,other,diagnostic=False): return CPerformanceIndicator.assertSame(self,other,[['rMeas',0.01,-1],['highResLimit',0.01,0],['spaceGroup','',0]],diagnostic=diagnostic) class CAtomCountPerformance(CPerformanceIndicator): # Trivial performance test for chainsaw/sculptor for use in project-based-testing CONTENTS_ORDER = [ 'nAtoms', 'nResidues'] CONTENTS = { 'nAtoms' : { 'class' : CCP4Data.CInt }, 'nResidues' : { 'class' : CCP4Data.CInt } } def assertSame(self,other,diagnostic=False): return CPerformanceIndicator.assertSame(self,other,[ ['nAtoms',0,1 ] ] ,diagnostic=diagnostic) def __str__(self): text = '' if self.__dict__['_value']['nResidues'].isSet(): text = text + 'nRes='+str(self.__dict__['_value']['nResidues']) elif self.__dict__['_value']['nAtoms'].isSet(): text = text + 'nAtoms='+str(self.__dict__['_value']['nAtoms']) return text def saveToDb(self): ret = {} if self.__dict__['_value']['nAtoms'].isSet(): ret['nAtoms'] = int(self.__dict__['_value']['nAtoms']) if self.__dict__['_value']['nResidues'].isSet(): ret['nResidues'] = int(self.__dict__['_value']['nResidues']) print 'CAtomCountPerformance.saveToDb',ret return [],None,ret def testComparisonData(self): ret = { } if self.__dict__['_value']['nAtoms'].isSet(): ret['nAtoms'] = self.__dict__['_value']['nAtoms'] if self.__dict__['_value']['nResidues'].isSet(): ret['nResidues'] = self.__dict__['_value']['nResidues'] return [],None,ret def setFromPdbDataFile(self,fileName): if isinstance(fileName,str): import CCP4ModelData fileName = CCP4ModelData.CPdbDataFile(fileName) try: fileName.loadFile() print 'CAtomCountPerformance.setFromPdbDataFile',fileName.fileContent.composition.nAtoms self.__dict__['_value']['nAtoms'].set(fileName.fileContent.composition.nAtoms) self.__dict__['_value']['nResidues'].set(fileName.fileContent.composition.nResidues) except: pass class CTestObsConversionsPerformance(CPerformanceIndicator): # Trivial performance test for chainsaw/sculptor for use in project-based-testing CONTENTS_ORDER = [ 'columnLabelsString'] CONTENTS = { 'columnLabelsString' : { 'class' : CCP4Data.CString } } def assertSame(self,other,diagnostic=False): return CPerformanceIndicator.assertSame(self,other,[ ['columnLabelsString',0,1 ] ] ,diagnostic=diagnostic) def __str__(self): text = '' if self.__dict__['_value']['columnLabelsString'].isSet(): text = text + 'labels:'+str(self.__dict__['_value']['columnLabelsString']).replace("'","") return text def saveToDb(self): ret = {} if self.__dict__['_value']['columnLabelsString'].isSet(): ret['columnLabelsString'] = str(self.__dict__['_value']['columnLabelsString']) print 'CTestObsConversionsPerformance.saveToDb',ret return [],None,ret def testComparisonData(self): ret = { } if self.__dict__['_value']['columnLabelsString'].isSet(): ret['columnLabelsString'] = self.__dict__['_value']['columnLabelsString'] return [],None,ret class CExpPhasPerformance(CPerformanceIndicator): CONTENTS_ORDER = ['FOM','CFOM','Hand1Score','Hand2Score','CC','RFactor','RFree','annotation'] CONTENTS = { 'FOM' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'CFOM' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'Hand1Score' : { 'class' : CCP4Data.CFloat }, 'Hand2Score' : { 'class' : CCP4Data.CFloat }, 'CC' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'RFactor' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'RFree' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'annotation' : { 'class' : CCP4Data.CString } } def __str__(self): text, val = '', self.__dict__['_value'] if val['FOM'].isSet(): text += 'FOM={:.2f} '.format(val['FOM']) if val['CFOM'].isSet(): text += 'CFOM={:.1f} '.format(val['CFOM']) if val['Hand1Score'].isSet() and val['Hand2Score'].isSet(): if val['CC'].isSet() and min(abs(val['CC']-val['Hand1Score']),abs(val['CC']-val['Hand2Score']))<1e-4: text += 'CC={:.0f}:{:.0f} '.format(val['Hand1Score'],val['Hand2Score']) else: text += 'Score={:.0f}:{:.0f} '.format(val['Hand1Score'],val['Hand2Score']) elif val['CC'].isSet(): text += 'CC={:.1f} '.format(val['CC']) if val['RFactor'].isSet(): text += 'R={:.2f} '.format(val['RFactor']) if val['RFree'].isSet(): text += 'Rfree={:.2f} '.format(val['RFree']) return text def saveToDb(self): #Return a list of files, string representation of xml, dict of key,value pairs #The dict keys must map to those defined in CCP4DbApi. ret = {} if self.FOM.isSet(): ret['FOM'] = self.FOM.__float__() if self.CFOM.isSet(): ret['CFOM'] = self.CFOM.__float__() if self.Hand1Score.isSet(): ret['Hand1Score'] = self.Hand1Score.__float__() if self.Hand2Score.isSet(): ret['Hand2Score'] = self.Hand2Score.__float__() if self.CC.isSet(): ret['CC'] = self.CC.__float__() if self.RFactor.isSet(): ret['RFactor'] = self.RFactor.__float__() if self.RFree.isSet(): ret['RFree'] = self.RFree.__float__() return [],None,ret def assertSame(self,other,diagnostic=False): return CPerformanceIndicator.assertSame(self,other,[['FOM',0.05,-1],['CFOM',10.,-1],['Hand1Score',10.,-1],['Hand2Score',10.,-1],['CC',0.01,-1],['RFactor',0.01,-1],['RFree',0.01,-1],],diagnostic=diagnostic) class CPhaseErrorPerformance(CPerformanceIndicator): CONTENTS_ORDER = ['phaseError','weightedPhaseError','reflectionCorrelation'] CONTENTS = { 'phaseError' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'weightedPhaseError' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } }, 'reflectionCorrelation' : { 'class' : CCP4Data.CFloat, 'qualifiers' : { 'min' : 0.0 } } } def __str__(self): #if self.__dict__['_value']['phaseError'].__dict__['_value'] is not None: text = 'dphi=' + format(self.__dict__['_value']['phaseError'],'.1f') return text def saveToDb(self): #Return a list of files, string reporesentation of xml, dict of key,value pairs #The dict keys must map to does defined in CCP4DbApi. ret = {} if self.phaseError.isSet(): ret['phaseError'] = self.phaseError.__float__() if self.weightedPhaseError.isSet(): ret['weightedPhaseError'] = self.weightedPhaseError.__float__() if self.reflectionCorrelation.isSet(): ret['reflectionCorrelation'] = self.reflectionCorrelation.__float__() return [],None,ret def assertSame(self,other,diagnostic=False): return CPerformanceIndicator.assertSame(self,other,[['phaseError',5.0,-1],['weightedPhaseError',5.0,-1],['reflectionCorrelation',0.05,1]],diagnostic=diagnostic) class CSuperposePerformance(CPerformanceIndicator): # Trivial performance test for chainsaw/sculptor for use in project-based-testing CONTENTS_ORDER = [ 'RMSxyz' ,'nResidues' ] CONTENTS = { 'RMSxyz' : { 'class' : CCP4Data.CFloat } , 'nResidues' : { 'class' : CCP4Data.CInt } } # 'QScore' : { 'class' : CCP4Data.CFloat }, def assertSame(self,other,diagnostic=False): #return CPerformanceIndicator.assertSame(self,other,[ ['nResidues',0,1 ], [ 'QScore', 1.0, 1], ['RMSxyz', 0.1, -1 ] ] ,diagnostic=diagnostic) return CPerformanceIndicator.assertSame(self,other,[ ['RMSxyz', 0.1, -1 ], ['nResidues',0,1 ]] ,diagnostic=diagnostic) def __str__(self): text = '' if self.__dict__['_value']['RMSxyz'].isSet(): text = text + 'RMS='+str(self.__dict__['_value']['RMSxyz']) + ' ' if self.__dict__['_value']['nResidues'].isSet(): text = text + 'nRes='+str(self.__dict__['_value']['nResidues']) + ' ' return text def saveToDb(self): ret = {} if self.__dict__['_value']['RMSxyz'].isSet(): ret['RMSxyz'] = float(self.__dict__['_value']['RMSxyz']) if self.__dict__['_value']['nResidues'].isSet(): ret['nResidues'] = int(self.__dict__['_value']['nResidues']) #if self.__dict__['_value']['QScore'].isSet(): ret['QScore'] = float(self.__dict__['_value']['QScore']) return [],None,ret