def reporter(self):
"""
Creates the metadata report by pulling specific attributes from the metadata objects
"""
printtime('Creating summary report', self.starttime)
header = '{}\n'.format(','.join(self.headers))
# Create a string to store all the results
data = str()
for sample in self.metadata:
# Add the value of the appropriate attribute to the results string
data += GenObject.returnattr(sample, 'name')
# SampleName
data += GenObject.returnattr(sample.run, 'SamplePlate')
# Genus
data += GenObject.returnattr(sample.sixteens_full, 'genus')
# SequencingDate
data += GenObject.returnattr(sample.run, 'Date')
# Analyst
data += GenObject.returnattr(sample.run, 'InvestigatorName')
# SamplePurity
data += GenObject.returnattr(sample.confindr, 'contam_status')
# GenomeQAML prediction
prediction = GenObject.returnattr(sample.GenomeQAML, 'prediction')
if prediction != ',':
data += prediction
else:
try:
description = sample.run.Description
if description == 'metagenome':
data += '{description},'.format(
description=description)
else:
data += '{status},'.format(status=sample.run.status)
except KeyError:
data += 'ND,'
# N50
n50 = GenObject.returnattr(sample.quality_features_polished, 'n50')
if n50 != '-,':
data += n50
else:
data += 'ND,'
# NumContigs
data += GenObject.returnattr(sample.quality_features_polished,
'num_contigs')
# TotalLength
data += GenObject.returnattr(sample.quality_features_polished,
'genome_length')
# MeanInsertSize
data += GenObject.returnattr(sample.mapping, 'MeanInsertSize')
# InsertSizeSTD
data += GenObject.returnattr(sample.mapping, 'StdInsertSize')
# AverageCoverageDepth
data += GenObject.returnattr(sample.mapping, 'MeanCoveragedata')
# CoverageDepthSTD
data += GenObject.returnattr(sample.mapping, 'StdCoveragedata')
# PercentGC
data += GenObject.returnattr(sample.quality_features_polished,
'gc')
# MASH_ReferenceGenome
data += GenObject.returnattr(sample.mash, 'closestrefseq')
# MASH_NumMatchingHashes
data += GenObject.returnattr(sample.mash, 'nummatches')
# 16S_result
data += GenObject.returnattr(sample.sixteens_full,
'sixteens_match')
# rMLST_Result
try:
# If the number of matches to the closest reference profile is 53, return the profile number
if sample.rmlst.matches == 53:
data += GenObject.returnattr(sample.rmlst, 'sequencetype')
else:
# Create a set of all the genes present in the results (gene name split from allele)
rmlst_gene_set = {
gene.split('_')[0]
for gene in sample.rmlst.results
}
# If there are a full set of 53 genes, but no profile match, then this is a new profile
if len(rmlst_gene_set) == 53:
data += 'new,'
# Otherwise the profile is set to ND
else:
data += 'ND,'
except KeyError:
data += 'ND,'
# MLST_Result
try:
if sample.mlst.matches == 7:
data += GenObject.returnattr(sample.mlst, 'sequencetype')
else:
# Create a set of all the genes present in the results (gene name split from allele)
mlst_gene_set = {
gene.split('_')[0]
for gene in sample.mlst.results
}
# If there are all the genes present, but no perfect match to a reference profile, state that
# the profile is new
if len(mlst_gene_set) == 7:
data += 'new,'
# Otherwise indicate that the profile is ND
else:
data += 'ND,'
except KeyError:
data += 'ND,'
# MLST_gene_X_alleles
try:
# Create a set of all the genes present in the results (gene name split from allele)
gene_set = {gene.split('_')[0] for gene in sample.mlst.results}
for gene in sorted(gene_set):
allele_list = list()
# Determine all the alleles that are present for each gene
for allele in sample.mlst.results:
if gene in allele:
allele_list.append(allele)
# If there is more than one allele in the sample, add both to the string separated by a ';'
if len(allele_list) > 1:
data += '{},'.format(';'.join(allele_list))
# Otherwise add the only allele
else:
data += allele_list[0] + ','
# If there are fewer than seven matching alleles, add a ND for each missing result
if len(gene_set) < 7:
data += (7 - len(gene_set)) * 'ND,'
except KeyError:
# data += '-,-,-,-,-,-,-,'
data += 'ND,ND,ND,ND,ND,ND,ND,'
# CoreGenesPresent
data += GenObject.returnattr(sample.coregenome, 'coreresults')
# E_coli_Serotype
try:
serotype = '{oset} ({opid}):{hset} ({hpid}),'\
.format(oset=';'.join(sample.serosippr.o_set),
opid=sample.serosippr.best_o_pid,
hset=';'.join(sample.serosippr.h_set),
hpid=sample.serosippr.best_h_pid)
# Make sure that the string was populated with values rather than 'NA' or '-'
if serotype == '- (-):- (-),':
data += 'ND,'
else:
data += serotype
except KeyError:
data += 'ND,'
# SISTR_serovar_antigen
data += GenObject.returnattr(sample.sistr,
'serovar_antigen').rstrip(';')
# SISTR_serovar_cgMLST
data += GenObject.returnattr(sample.sistr, 'serovar_cgmlst')
# SISTR_serogroup
data += GenObject.returnattr(sample.sistr, 'serogroup')
# SISTR_h1
data += GenObject.returnattr(sample.sistr, 'h1').rstrip(';')
# SISTR_h2
data += GenObject.returnattr(sample.sistr, 'h2').rstrip(';')
# SISTR_serovar
data += GenObject.returnattr(sample.sistr, 'serovar')
# GeneSeekr_Profile
try:
if sample.genesippr.report_output:
data += ';'.join(sample.genesippr.report_output) + ','
else:
data += 'ND,'
except KeyError:
data += 'ND,'
# Vtyper_Profile
try:
# Since the vtyper attribute can be empty, check first
profile = sorted(sample.vtyper.profile)
if profile:
data += ';'.join(profile) + ','
else:
data += 'ND,'
except KeyError:
data += 'ND,'
# AMR_Profile and resistant/sensitive status
if sample.resfinder_assembled.pipelineresults:
# Profile
data += ';'.join(
sorted(sample.resfinder_assembled.pipelineresults)) + ','
# Resistant/Sensitive
data += 'Resistant,'
else:
# Profile
data += 'ND,'
# Resistant/Sensitive
data += 'Sensitive,'
# Plasmid Result'
try:
plasmid_profile = sorted(sample.plasmidextractor.plasmids)
if plasmid_profile:
data += ';'.join(plasmid_profile) + ','
else:
data += 'ND,'
except KeyError:
data += 'ND,'
# TotalPredictedGenes
data += GenObject.returnattr(sample.prodigal,
'predictedgenestotal')
# PredictedGenesOver3000bp
data += GenObject.returnattr(sample.prodigal,
'predictedgenesover3000bp')
# PredictedGenesOver1000bp
data += GenObject.returnattr(sample.prodigal,
'predictedgenesover1000bp')
# PredictedGenesOver500bp
data += GenObject.returnattr(sample.prodigal,
'predictedgenesover500bp')
# PredictedGenesUnder500bp
data += GenObject.returnattr(sample.prodigal,
'predictedgenesunder500bp')
# NumClustersPF
data += GenObject.returnattr(sample.run, 'NumberofClustersPF')
# Percent of reads mapping to PhiX control
data += GenObject.returnattr(sample.run, 'phix_aligned')
# Error rate calculated from PhiX control
data += GenObject.returnattr(sample.run, 'error_rate')
# LengthForwardRead
data += GenObject.returnattr(sample.run, 'forwardlength')
# LengthReverseRead
data += GenObject.returnattr(sample.run, 'reverselength')
# Real time strain
data += GenObject.returnattr(sample.run, 'Description')
# Flowcell
data += GenObject.returnattr(sample.run, 'flowcell')
# MachineName
data += GenObject.returnattr(sample.run, 'instrument')
# PipelineVersion
data += self.commit + ','
# AssemblyDate
data += datetime.now().strftime('%Y-%m-%d')
# Append a new line to the end of the results for this sample
data += '\n'
# Replace any NA values with -
cleandata = data.replace('NA', 'ND')
with open(os.path.join(self.reportpath, 'combinedMetadata.csv'),
'w') as metadatareport:
metadatareport.write(header)
metadatareport.write(cleandata)
def legacy_reporter(self):
"""
Creates an output that is compatible with the legacy metadata reports. This method will be removed once
a new database scheme is implemented
"""
from collections import OrderedDict
printtime('Creating legacy summary report', self.starttime)
row = ''
# Create a dictionary of tuples to be printed in the final report
for sample in self.metadata:
data = OrderedDict([
('SampleName', sample.name),
('N50', str(sample.quality_features_polished.n50)),
('NumContigs',
str(sample.quality_features_polished.num_contigs)),
('TotalLength',
str(sample.quality_features_polished.genome_length)),
('MeanInsertSize', sample.mapping.MeanInsertSize),
('AverageCoverageDepth',
sample.mapping.MeanCoveragedata.split("X")[0]),
('ReferenceGenome', sample.mash.closestrefseq),
('RefGenomeAlleleMatches', '-'),
('16sPhylogeny', sample.sixteens_full.genus),
('rMLSTsequenceType', sample.rmlst.sequencetype),
('MLSTsequencetype', sample.mlst.sequencetype),
('MLSTmatches', str(sample.mlst.matchestosequencetype)),
('coreGenome',
GenObject.returnattr(sample.coregenome,
'coreresults').rstrip(',')),
('SeroType', '{oset}:{hset}'.format(
oset=';'.join(sample.serosippr.o_set),
hset=';'.join(sample.serosippr.h_set))),
('geneSeekrProfile', ';'.join(result for result, pid in sorted(
sample.genesippr.results.items()))),
('vtyperProfile', ';'.join(sorted(sample.vtyper.profile))),
('percentGC', str(sample.quality_features_polished.gc)),
('TotalPredictedGenes',
str(sample.prodigal.predictedgenestotal)),
('predictedgenesover3000bp',
str(sample.prodigal.predictedgenesover3000bp)),
('predictedgenesover1000bp',
str(sample.prodigal.predictedgenesover1000bp)),
('predictedgenesover500bp',
str(sample.prodigal.predictedgenesover500bp)),
('predictedgenesunder500bp',
str(sample.prodigal.predictedgenesunder500bp)),
('SequencingDate', sample.run.Date),
('Investigator', sample.run.InvestigatorName),
('TotalClustersinRun', str(sample.run.TotalClustersinRun)),
('NumberofClustersPF', str(sample.run.NumberofClustersPF)),
('PercentOfClusters', str(sample.run.PercentOfClusters)),
('LengthofForwardRead', str(sample.run.forwardlength)),
('LengthofReverseRead', str(sample.run.reverselength)),
('Project', str(sample.run.SampleProject)),
('PipelineVersion', self.commit)
])
if not row:
row += ','.join([key for key, value in data.items()])
row += '\n'
row += ','.join([value for key, value in data.items()])
cleanrow = row.replace('NA', '').replace(',-,', ',,')
with open(os.path.join(self.reportpath, 'legacy_combinedMetadata.csv'),
'w') as metadatareport:
metadatareport.write(cleanrow)