def reportFullMSA(self, outDir, outFile):
"""Create MSA with all reference and bin alignments."""
# write bin alignments to file
oldStdOut = reassignStdOut(outFile)
for line in open(
os.path.join(outDir, 'storage', 'tree',
DefaultValues.PPLACER_CONCAT_SEQ_OUT)):
print((line.rstrip()))
# read duplicate seqs
duplicateNodes = self.__readDuplicateSeqs()
# write reference alignments to file
seqs = readFasta(
os.path.join(DefaultValues.PPLACER_REF_PACKAGE_FULL,
DefaultValues.GENOME_TREE_FASTA))
for seqId, seq in seqs.items():
print(('>' + seqId))
print(seq)
if seqId in duplicateNodes:
for dupSeqId in duplicateNodes[seqId]:
print(('>' + dupSeqId))
print(seq)
restoreStdOut(outFile, oldStdOut)
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 __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 reportNewickTree(self, outDir, outFile, leafLabels=None):
# read duplicate nodes
duplicateSeqs = self.__readDuplicateSeqs()
# read tree
treeFile = os.path.join(outDir, 'storage', 'tree',
DefaultValues.PPLACER_TREE_OUT)
tree = dendropy.Tree.get_from_path(treeFile,
schema='newick',
rooting="force-rooted",
preserve_underscores=True)
# clean up internal node labels
for node in tree.internal_nodes():
if node.label:
labelSplit = node.label.split('|')
label = labelSplit[0]
if labelSplit[1] != '':
label += '|' + labelSplit[1]
if labelSplit[2] != '':
label += '|' + labelSplit[2]
node.label = label
# insert duplicate nodes into tree
for leaf in tree.leaf_nodes():
duplicates = duplicateSeqs.get(leaf.taxon.label, None)
if duplicates != None:
newParent = leaf.parent_node.new_child(
edge_length=leaf.edge_length)
curLeaf = leaf.parent_node.remove_child(leaf)
newParent.new_child(taxon=curLeaf.taxon, edge_length=0)
for d in duplicates:
newParent.new_child(taxon=dendropy.Taxon(label=d),
edge_length=0)
# append taxonomy to leaf nodes
if leafLabels == 'taxonomy':
# read taxonomy string for each IMG genome
taxonomy = {}
for line in open(
os.path.join(DefaultValues.GENOME_TREE_DIR,
DefaultValues.GENOME_TREE_TAXONOMY)):
lineSplit = line.split('\t')
taxonomy[lineSplit[0]] = lineSplit[1].rstrip()
# append taxonomy to leaf labels
for leaf in tree.leaf_nodes():
taxaStr = taxonomy.get(leaf.taxon.label, None)
if taxaStr:
leaf.taxon.label += '|' + taxaStr
# write out tree
oldStdOut = reassignStdOut(outFile)
print((tree.as_string(schema='newick', suppress_rooting=True)))
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 reportNewickTree(self, outDir, outFile, leafLabels=None):
# read duplicate nodes
duplicateSeqs = self.__readDuplicateSeqs()
# read tree
treeFile = os.path.join(outDir, 'storage', 'tree', DefaultValues.PPLACER_TREE_OUT)
tree = dendropy.Tree.get_from_path(treeFile, schema='newick', as_rooted=True, preserve_underscores=True)
# clean up internal node labels
for node in tree.internal_nodes():
if node.label:
labelSplit = node.label.split('|')
label = labelSplit[0]
if labelSplit[1] != '':
label += '|' + labelSplit[1]
if labelSplit[2] != '':
label += '|' + labelSplit[2]
node.label = label
# insert duplicate nodes into tree
for leaf in tree.leaf_nodes():
duplicates = duplicateSeqs.get(leaf.taxon.label, None)
if duplicates != None:
newParent = leaf.parent_node.new_child(edge_length = leaf.edge_length)
curLeaf = leaf.parent_node.remove_child(leaf)
newParent.new_child(taxon = curLeaf.taxon, edge_length = 0)
for d in duplicates:
newParent.new_child(taxon = Taxon(label = d), edge_length = 0)
# append taxonomy to leaf nodes
if leafLabels == 'taxonomy':
# read taxonomy string for each IMG genome
taxonomy = {}
for line in open(os.path.join(DefaultValues.GENOME_TREE_DIR, 'genome_tree.taxonomy.tsv')):
lineSplit = line.split('\t')
taxonomy[lineSplit[0]] = lineSplit[1].rstrip()
# append taxonomy to leaf labels
for leaf in tree.leaf_nodes():
taxaStr = taxonomy.get(leaf.taxon.label, None)
if taxaStr:
leaf.taxon.label += '|' + taxaStr
# write out tree
oldStdOut = reassignStdOut(outFile)
print(tree.as_string(schema='newick', suppress_rooting=True))
restoreStdOut(outFile, oldStdOut)
def joinTables(self, options):
"""Join tables command"""
self.logger.info(
'[CheckM - join_tables] Joining tables containing bin information.'
)
# read all tables
headers = {}
rows = defaultdict(dict)
binIds = set()
for f in options.tables:
with open(f) as fin:
headers[f] = [x.strip()
for x in fin.readline().split('\t')][1:]
for line in fin:
lineSplit = [x.strip() for x in line.split('\t')]
binId = lineSplit[0]
binIds.add(binId)
for i, header in enumerate(headers[f]):
rows[binId][header] = lineSplit[i + 1]
# write merge table
oldStdOut = reassignStdOut(options.file)
row = 'Bin Id'
for f in options.tables:
row += '\t' + '\t'.join(headers[f])
print(row)
for binId in binIds:
row = binId
for f in options.tables:
for header in headers[f]:
row += '\t' + rows[binId].get(header, '')
print(row)
restoreStdOut(options.file, oldStdOut)
if options.file:
self.logger.info('Joined table written to: ' + options.file)
self.timeKeeper.printTimeStamp()
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[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 reportFullMSA(self, outDir, outFile):
"""Create MSA with all reference and bin alignments."""
# write bin alignments to file
oldStdOut = reassignStdOut(outFile)
for line in open(os.path.join(outDir, 'storage', 'tree', DefaultValues.PPLACER_CONCAT_SEQ_OUT)):
print(line.rstrip())
# read duplicate seqs
duplicateNodes = self.__readDuplicateSeqs()
# write reference alignments to file
seqs = readFasta(os.path.join(DefaultValues.PPLACER_REF_PACKAGE, 'genome_tree.concatenated.derep.fasta'))
for seqId, seq in seqs.iteritems():
print('>' + seqId)
print(seq)
if seqId in duplicateNodes:
for dupSeqId in duplicateNodes[seqId]:
print('>' + dupSeqId)
print(seq)
restoreStdOut(outFile, oldStdOut)
def printSummary(self, outputFormat, aai, binIdToBinMarkerSets, bIndividualMarkers, coverageFile, bTabTable, outFile):
# redirect output
oldStdOut = reassignStdOut(outFile)
coverageBinProfiles = None
if coverageFile:
coverage = Coverage(1)
coverageBinProfiles = coverage.binProfiles(coverageFile)
prettyTableFormats = [1, 2, 3]
header = self.__getHeader(outputFormat, binIdToBinMarkerSets[binIdToBinMarkerSets.keys()[0]], coverageBinProfiles)
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
for binId in sorted(self.results.keys()):
self.results[binId].printSummary(outputFormat, aai, binIdToBinMarkerSets[binId], bIndividualMarkers, coverageBinProfiles, pTable)
if not bTabTable :
if outputFormat in [1,2]:
print(pTable.get_string(sortby='Completeness', reversesort=True))
else:
print(pTable.get_string())
# restore stdout
restoreStdOut(outFile, oldStdOut)
def run(self, binFiles, bamFiles, outFile, bAllReads, minAlignPer,
maxEditDistPer, minQC):
"""Calculate coverage of sequences for each BAM file."""
# determine bin assignment of each sequence
self.logger.info(' Determining bin assignment of each sequence.')
seqIdToBinId = {}
seqIdToSeqLen = {}
for binFile in binFiles:
binId = binIdFromFilename(binFile)
seqs = readFasta(binFile)
for seqId, seq in seqs.iteritems():
seqIdToBinId[seqId] = binId
seqIdToSeqLen[seqId] = len(seq)
# process each fasta file
self.logger.info(" Processing %d file(s) with %d threads.\n" %
(len(bamFiles), self.totalThreads))
# make sure all BAM files are sorted
self.numFiles = len(bamFiles)
for bamFile in bamFiles:
if not os.path.exists(bamFile + '.bai'):
self.logger.error(
' [Error] BAM file is either unsorted or not indexed: ' +
bamFile + '\n')
sys.exit()
# calculate coverage of each BAM file
coverageInfo = {}
numFilesStarted = 0
for bamFile in bamFiles:
numFilesStarted += 1
self.logger.info(
' Processing %s (%d of %d):' %
(ntpath.basename(bamFile), numFilesStarted, len(bamFiles)))
coverageInfo[bamFile] = mp.Manager().dict()
coverageInfo[bamFile] = self.__processBam(bamFile, bAllReads,
minAlignPer,
maxEditDistPer, minQC,
coverageInfo[bamFile])
# redirect output
self.logger.info(' Writing coverage information to file.')
oldStdOut = reassignStdOut(outFile)
header = 'Sequence Id\tBin Id\tSequence length (bp)'
for bamFile in bamFiles:
header += '\tBam Id\tCoverage\tMapped reads'
print(header)
# get length of all seqs
for bamFile, seqIds in coverageInfo.iteritems():
for seqId in seqIds.keys():
seqIdToSeqLen[seqId] = seqIds[seqId].seqLen
# write coverage stats for all scaffolds to file
for seqId, seqLen in seqIdToSeqLen.iteritems():
rowStr = seqId + '\t' + seqIdToBinId.get(
seqId, DefaultValues.UNBINNED) + '\t' + str(seqLen)
for bamFile in bamFiles:
bamId = binIdFromFilename(bamFile)
if seqId in coverageInfo[bamFile]:
rowStr += '\t%s\t%f\t%d' % (
bamId, coverageInfo[bamFile][seqId].coverage,
coverageInfo[bamFile][seqId].mappedReads)
else:
rowStr += '\t%s\t%f\t%d' % (bamId, 0, 0)
print(rowStr)
# restore stdout
restoreStdOut(outFile, oldStdOut)
def run(self, binFiles, bamFiles, outFile, bAllReads, minAlignPer, maxEditDistPer, minQC):
"""Calculate coverage of sequences for each BAM file."""
# determine bin assignment of each sequence
self.logger.info(' Determining bin assignment of each sequence.')
seqIdToBinId = {}
seqIdToSeqLen = {}
for binFile in binFiles:
binId = binIdFromFilename(binFile)
seqs = readFasta(binFile)
for seqId, seq in seqs.iteritems():
seqIdToBinId[seqId] = binId
seqIdToSeqLen[seqId] = len(seq)
# process each fasta file
self.logger.info(" Processing %d file(s) with %d threads.\n" % (len(bamFiles), self.totalThreads))
# make sure all BAM files are sorted
self.numFiles = len(bamFiles)
for bamFile in bamFiles:
if not os.path.exists(bamFile + '.bai'):
self.logger.error(' [Error] BAM file is either unsorted or not indexed: ' + bamFile + '\n')
sys.exit(1)
# calculate coverage of each BAM file
coverageInfo = {}
numFilesStarted = 0
for bamFile in bamFiles:
numFilesStarted += 1
self.logger.info(' Processing %s (%d of %d):' % (ntpath.basename(bamFile), numFilesStarted, len(bamFiles)))
coverageInfo[bamFile] = mp.Manager().dict()
coverageInfo[bamFile] = self.__processBam(bamFile, bAllReads, minAlignPer, maxEditDistPer, minQC, coverageInfo[bamFile])
# redirect output
self.logger.info(' Writing coverage information to file.')
oldStdOut = reassignStdOut(outFile)
header = 'Sequence Id\tBin Id\tSequence length (bp)'
for bamFile in bamFiles:
header += '\tBam Id\tCoverage\tMapped reads'
print(header)
# get length of all seqs
for bamFile, seqIds in coverageInfo.iteritems():
for seqId in seqIds.keys():
seqIdToSeqLen[seqId] = seqIds[seqId].seqLen
# write coverage stats for all scaffolds to file
for seqId, seqLen in seqIdToSeqLen.iteritems():
rowStr = seqId + '\t' + seqIdToBinId.get(seqId, DefaultValues.UNBINNED) + '\t' + str(seqLen)
for bamFile in bamFiles:
bamId = binIdFromFilename(bamFile)
if seqId in coverageInfo[bamFile]:
rowStr += '\t%s\t%f\t%d' % (bamId, coverageInfo[bamFile][seqId].coverage, coverageInfo[bamFile][seqId].mappedReads)
else:
rowStr += '\t%s\t%f\t%d' % (bamId, 0, 0)
print(rowStr)
# 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 __printFullTable(self, binIdToTaxonomy, binIdToSisterTaxonomy, binIdToLineageStatistics, resultsParser, binStats, 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"]
header += ['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['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 += [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)
