def list(self, rankFilter='ALL'):
""" List all available marker sets from the specified rank."""
taxonMarkerSets = self.readMarkerSets()
header = [
'Rank', 'Taxon', '# genomes', '# marker genes', '# marker sets'
]
pTable = prettytable.PrettyTable(header)
pTable.align = 'c'
pTable.align['Rank'] = 'l'
pTable.align['Taxon'] = 'l'
pTable.hrules = prettytable.FRAME
pTable.vrules = prettytable.NONE
for rank in taxonomicRanks:
if rankFilter == 'ALL' or rankFilter == rank:
for taxon in sorted(taxonMarkerSets[rank]):
markerSet = taxonMarkerSets[rank][taxon]
numMarkers, numMarkerSets = markerSet.size()
pTable.add_row([
rank, taxon, markerSet.numGenomes, numMarkers,
numMarkerSets
])
print('')
print(pTable.get_string())
def __printSimpleSummaryTable(self, binIdToTaxonomy, resultsParser, bTabTable, outFile):
# redirect output
oldStdOut = reassignStdOut(outFile)
arbitraryBinId = binIdToTaxonomy.keys()[0]
markerCountLabel = '# unique markers (of %d)' % len(resultsParser.models[arbitraryBinId])
header = ['Bin Id', markerCountLabel, '# multi-copy', 'Taxonomy']
if bTabTable:
pTable = None
print('\t'.join(header))
else:
pTable = prettytable.PrettyTable(header)
pTable.float_format = '.2'
pTable.align = 'c'
pTable.align[header[0]] = 'l'
pTable.align['Taxonomy'] = 'l'
pTable.hrules = prettytable.FRAME
pTable.vrules = prettytable.NONE
for binId in sorted(binIdToTaxonomy.keys()):
uniqueHits, multiCopyHits = resultsParser.results[binId].countUniqueHits()
row = [binId, uniqueHits, multiCopyHits, binIdToTaxonomy[binId]]
if bTabTable:
print('\t'.join(map(str, row)))
else:
pTable.add_row(row)
if not bTabTable:
print(pTable.get_string(sortby=markerCountLabel, reversesort=True))
# restore stdout
restoreStdOut(outFile, oldStdOut)
def printSummary(self, outputFormat, aai, binIdToBinMarkerSets,
bIndividualMarkers, coverageFile, bTabTable, outFile,
anaFolder):
# redirect output
oldStdOut = reassignStdOut(outFile)
coverageBinProfiles = None
if coverageFile:
coverage = Coverage(1)
coverageBinProfiles = coverage.binProfiles(coverageFile)
prettyTableFormats = [1, 2, 3, 9]
header = self.__getHeader(
outputFormat,
binIdToBinMarkerSets[list(binIdToBinMarkerSets.keys())[0]],
coverageBinProfiles, bTabTable)
if bTabTable or outputFormat not in prettyTableFormats:
bTabTable = True
pTable = None
if header != None:
print('\t'.join(header))
else:
pTable = prettytable.PrettyTable(header)
pTable.float_format = '.2'
pTable.align = 'c'
pTable.align[header[0]] = 'l'
pTable.hrules = prettytable.FRAME
pTable.vrules = prettytable.NONE
seqsReported = 0
for binId in sorted(self.results.keys()):
seqsReported += self.results[binId].printSummary(
outputFormat, aai, binIdToBinMarkerSets[binId],
bIndividualMarkers, coverageBinProfiles, pTable, anaFolder)
if outputFormat in [6, 7] and seqsReported == 0:
print('[No marker genes satisfied the reporting criteria.]')
if not bTabTable:
if outputFormat in [1, 2]:
print(
pTable.get_string(sortby='Completeness', reversesort=True))
else:
# only print if there are rows
if pTable.get_string(print_empty=False):
print(pTable.get_string(print_empty=False))
# restore stdout
restoreStdOut(outFile, oldStdOut)
def __printFullTable(self, binIdToUID, binIdToTaxonomy,
binIdToSisterTaxonomy, binIdToLineageStatistics,
resultsParser, binStats, bTabTable, outFile):
# redirect output
oldStdOut = reassignStdOut(outFile)
arbitraryBinId = list(binIdToTaxonomy.keys())[0]
markerCountLabel = '# unique markers (of %d)' % len(
resultsParser.models[arbitraryBinId])
header = ['Bin Id', markerCountLabel, "# multi-copy"]
header += [
'Insertion branch UID', 'Taxonomy (contained)',
'Taxonomy (sister lineage)'
]
header += [
'GC', 'Genome size (Mbp)', 'Gene count', 'Coding density',
'Translation table'
]
header += [
'# descendant genomes', 'Lineage: GC mean', 'Lineage: GC std'
]
header += [
'Lineage: genome size (Mbp) mean', 'Lineage: genome size (Mbp) std'
]
header += ['Lineage: gene count mean', 'Lineage: gene count std']
if bTabTable:
pTable = None
print(('\t'.join(header)))
else:
pTable = prettytable.PrettyTable(header)
pTable.float_format = '.2'
pTable.float_format['GC'] = '.1'
pTable.float_format['Lineage: GC mean'] = '.1'
pTable.float_format['Lineage: GC std'] = '.1'
pTable.float_format['Lineage: gene count mean'] = '.0'
pTable.float_format['Lineage: gene count std'] = '.0'
pTable.align = 'c'
pTable.align[header[0]] = 'l'
pTable.align['Insertion branch UID'] = 'l'
pTable.align['Taxonomy (contained)'] = 'l'
pTable.align['Taxonomy (sister lineage)'] = 'l'
pTable.hrules = prettytable.FRAME
pTable.vrules = prettytable.NONE
for binId in sorted(binIdToTaxonomy.keys()):
uniqueHits, multiCopyHits = resultsParser.results[
binId].countUniqueHits()
truncSisterLineage = binIdToSisterTaxonomy[binId]
for taxa in binIdToTaxonomy[binId].split(';'):
truncSisterLineage = truncSisterLineage.replace(taxa + ';', '')
if len(truncSisterLineage) == 0:
truncSisterLineage = 'unresolved'
elif truncSisterLineage[-1] == ';':
truncSisterLineage = truncSisterLineage[0:-1]
row = [binId, uniqueHits, multiCopyHits]
row += [
binIdToUID[binId], binIdToTaxonomy[binId], truncSisterLineage
]
row += [binStats[binId]['GC'] * 100]
row += [float(binStats[binId]['Genome size']) / 1e6]
row += [binStats[binId]['# predicted genes']]
row += [binStats[binId]['Coding density']]
row += [binStats[binId]['Translation table']]
row += [binIdToLineageStatistics[binId]['# genomes']]
row += [binIdToLineageStatistics[binId]['gc mean']]
row += [binIdToLineageStatistics[binId]['gc std']]
row += [binIdToLineageStatistics[binId]['genome size mean']]
row += [binIdToLineageStatistics[binId]['genome size std']]
row += [binIdToLineageStatistics[binId]['gene count mean']]
row += [binIdToLineageStatistics[binId]['gene count std']]
if bTabTable:
print(('\t'.join(map(str, row))))
else:
pTable.add_row(row)
if not bTabTable:
print((pTable.get_string(sortby=markerCountLabel,
reversesort=True)))
# restore stdout
restoreStdOut(outFile, oldStdOut)
def run(self, coverageFile, outFile, bTabTable):
checkFileExists(coverageFile)
# get number of reads mapped to each bin
self.logger.info('Determining number of reads mapped to each bin.')
readsMappedToBin = {}
binSize = {}
totalMappedReads = {}
bHeader = True
for line in open(coverageFile):
if bHeader:
bHeader = False
continue
lineSplit = line.split('\t')
# seqId = lineSplit[0]
binId = lineSplit[1]
seqLen = int(lineSplit[2])
binSize[binId] = binSize.get(binId, 0) + seqLen
if binId not in readsMappedToBin:
readsMappedToBin[binId] = {}
for i in range(3, len(lineSplit), 3):
bamId = lineSplit[i]
mappedReads = int(lineSplit[i + 2])
totalMappedReads[bamId] = totalMappedReads.get(bamId,
0) + mappedReads
readsMappedToBin[binId][bamId] = readsMappedToBin[binId].get(
bamId, 0) + mappedReads
# calculate percentage of mapped reads to binned populations
perMappedReads = {}
normBinCoverage = {}
sumNormBinCoverage = {}
for binId, bamIds in readsMappedToBin.items():
perMappedReads[binId] = {}
normBinCoverage[binId] = {}
for bamId in bamIds:
perMR = float(
readsMappedToBin[binId][bamId]) / totalMappedReads[bamId]
perMappedReads[binId][bamId] = perMR
if binId == DefaultValues.UNBINNED:
continue
normCoverage = perMR / binSize[binId]
normBinCoverage[binId][bamId] = normCoverage
sumNormBinCoverage[bamId] = sumNormBinCoverage.get(
bamId, 0) + normCoverage
for binId, bamIds in normBinCoverage.items():
for bamId in bamIds:
if sumNormBinCoverage[bamId] != 0:
normBinCoverage[binId][bamId] /= sumNormBinCoverage[bamId]
else:
normBinCoverage[binId][bamId] = 0
# write community profile
oldStdOut = reassignStdOut(outFile)
sortedBinIds = sorted(readsMappedToBin.keys())
sortedBamIds = sorted(readsMappedToBin[sortedBinIds[0]].keys())
header = ['Bin Id', 'Bin size (Mbp)']
for bamId in sortedBamIds:
header += [bamId + ': mapped reads']
header += [bamId + ': % mapped reads']
header += [bamId + ': % binned populations']
header += [bamId + ': % community']
if bTabTable:
print('\t'.join(header))
else:
pTable = prettytable.PrettyTable(header)
pTable.float_format = '.2'
pTable.align = 'c'
pTable.align[header[0]] = 'l'
pTable.hrules = prettytable.FRAME
pTable.vrules = prettytable.NONE
for binId in sortedBinIds:
row = [binId]
row += [float(binSize[binId]) / 1e6]
for bamId in sortedBamIds:
row += [readsMappedToBin[binId][bamId]]
row += [perMappedReads[binId][bamId] * 100.0]
if DefaultValues.UNBINNED in perMappedReads:
unbinnedPercentage = perMappedReads[
DefaultValues.UNBINNED][bamId]
else:
unbinnedPercentage = 0
if binId == DefaultValues.UNBINNED:
row += ['NA']
row += [unbinnedPercentage * 100.0]
else:
row += [normBinCoverage[binId][bamId] * 100.0]
row += [
normBinCoverage[binId][bamId] * 100.0 *
(1.0 - unbinnedPercentage)
]
if bTabTable:
print('\t'.join(list(map(str, row))))
else:
pTable.add_row(row)
if not bTabTable:
print(pTable.get_string())
restoreStdOut(outFile, oldStdOut)
