def outputFinalData(self, outputFname, key2dataLs=None, delimiter=None, header=None): """ 2013.07.18 header output is not dependent on key2dataLs anymore 2013.3.3 bugfix , added openMode='w' for MatrixFile() 2013.2.12 replace csv.writer with MatrixFile 2012.7.30 open the outputFname regardless whether there is data or not. 2012.1.9 """ writer = MatrixFile(inputFname=outputFname, delimiter=delimiter, openMode='w') if header and delimiter: writer.writerow(header) if key2dataLs and delimiter: keyLs = key2dataLs.keys() keyLs.sort() for key in keyLs: dataLs = key2dataLs.get(key) writer.writerow(list(key) + dataLs) writer.close()
def run(self): """ """ if self.debug: import pdb pdb.set_trace() reader = MatrixFile(inputFname=self.inputFname) reader.constructColName2IndexFromHeader() meanMendelErrorIndex = reader.getColIndexGivenColHeader( "meanMendelError") noOfLociIndex = reader.getColIndexGivenColHeader("sampled_base_count") sumOfMendelErrorIndex = reader.getColIndexGivenColHeader( "sumOfMendelError") plinkPedigreeFile = PlinkPedigreeFile(inputFname=self.pedigreeFname) familyStructureData = plinkPedigreeFile.getFamilyStructurePlinkWay() twoParentFamilyCountData = self.getNoOfFamiliesAndKidsGivenParentSetSize(noOfParents2FamilyData=familyStructureData.noOfParents2FamilyData, \ parentSetSize=2) singleParentFamilyCountData = self.getNoOfFamiliesAndKidsGivenParentSetSize(noOfParents2FamilyData=familyStructureData.noOfParents2FamilyData, \ parentSetSize=1) zeroParentFamilyCountData = self.getNoOfFamiliesAndKidsGivenParentSetSize(noOfParents2FamilyData=familyStructureData.noOfParents2FamilyData, \ parentSetSize=0) writer = MatrixFile(self.outputFname, openMode='w', delimiter='\t') header = ["ID", "noOfTotalLoci", \ "noOfTwoParentFamilies", "noOfParentsInTwoParentFamilies", "noOfKidsInTwoParentFamilies", "noOfIndividualsInTwoParentFamilies", \ "noOfSingleParentFamilies", "noOfParentsInSingleParentFamilies", "noOfKidsInSingleParentFamilies", "noOfIndividualsInSingleParentFamilies", \ "noOfZeroParentFamilies", "noOfParentsInZeroParentFamilies", "noOfKidsInZeroParentFamilies", "noOfIndividualsInZeroParentFamilies", \ "noOfTotalMendelErrors", \ "noOfMendelErrorsPerLocusPerNuclearFamily", "noOfMendelErrorsPerNuclearFamily"] writer.writeHeader(header) for row in reader: meanMendelError = float(row[meanMendelErrorIndex]) noOfLoci = int(row[noOfLociIndex]) sumOfMendelError = int(row[sumOfMendelErrorIndex]) noOfNuclearFamilies = twoParentFamilyCountData.noOfFamilies if noOfNuclearFamilies > 0: noOfMendelErrorsPerLocusPerNuclearFamily = meanMendelError / float( noOfNuclearFamilies) noOfMendelErrorsPerNuclearFamily = sumOfMendelError / float( noOfNuclearFamilies) else: noOfMendelErrorsPerLocusPerNuclearFamily = -1 noOfMendelErrorsPerNuclearFamily = -1 data_row = [row[0], noOfLoci, \ noOfNuclearFamilies, twoParentFamilyCountData.noOfParents, twoParentFamilyCountData.noOfKids, \ twoParentFamilyCountData.noOfIndividuals,\ singleParentFamilyCountData.noOfFamilies, singleParentFamilyCountData.noOfParents, singleParentFamilyCountData.noOfKids,\ singleParentFamilyCountData.noOfIndividuals,\ zeroParentFamilyCountData.noOfFamilies, zeroParentFamilyCountData.noOfParents, zeroParentFamilyCountData.noOfKids,\ zeroParentFamilyCountData.noOfIndividuals,\ sumOfMendelError, \ noOfMendelErrorsPerLocusPerNuclearFamily,noOfMendelErrorsPerNuclearFamily ] writer.writerow(data_row) plinkPedigreeFile.close() reader.close() writer.close()