def run(self, ubiquityThreshold, singleCopyThreshold, minGenomes, mostSpecificRank, minMarkers, completenessThreshold, contaminationThreshold):
print 'Ubiquity threshold: ' + str(ubiquityThreshold)
print 'Single-copy threshold: ' + str(singleCopyThreshold)
print 'Min. genomes: ' + str(minGenomes)
print 'Most specific taxonomic rank: ' + str(mostSpecificRank)
print 'Min markers: ' + str(minMarkers)
print 'Completeness threshold: ' + str(completenessThreshold)
print 'Contamination threshold: ' + str(contaminationThreshold)
img = IMG()
markerset = MarkerSet()
lineages = img.lineagesByCriteria(minGenomes, mostSpecificRank)
degenerateGenomes = {}
for lineage in lineages:
genomeIds = img.genomeIdsByTaxonomy(lineage, 'Final')
print ''
print 'Lineage ' + lineage + ' contains ' + str(len(genomeIds)) + ' genomes.'
# get table of PFAMs and do some initial filtering to remove PFAMs that are
# clearly not going to pass the ubiquity and single-copy thresholds
countTable = img.countTable(genomeIds)
countTable = img.filterTable(genomeIds, countTable, ubiquityThreshold*0.9, singleCopyThreshold*0.9)
markerGenes = markerset.markerGenes(genomeIds, countTable, ubiquityThreshold*len(genomeIds), singleCopyThreshold*len(genomeIds))
if len(markerGenes) < minMarkers:
continue
geneDistTable = img.geneDistTable(genomeIds, markerGenes, spacingBetweenContigs=1e6)
colocatedGenes = markerset.colocatedGenes(geneDistTable)
colocatedSets = markerset.colocatedSets(colocatedGenes, markerGenes)
for genomeId in genomeIds:
completeness, contamination = markerset.genomeCheck(colocatedSets, genomeId, countTable)
if completeness < completenessThreshold or contamination > contaminationThreshold:
degenerateGenomes[genomeId] = degenerateGenomes.get(genomeId, []) + [[lineage.split(';')[-1].strip(), len(genomeIds), len(colocatedSets), completeness, contamination]]
# write out degenerate genomes
metadata = img.genomeMetadata('Final')
fout = open('./data/degenerate_genomes.tsv', 'w')
fout.write('Genome Id\tTaxonomy\tGenome Size (Gbps)\tScaffolds\tBiotic Relationships\tStatus\tLineage\t# genomes\tMarker set size\tCompleteness\tContamination\n')
for genomeId, data in degenerateGenomes.iteritems():
fout.write(genomeId + '\t' + '; '.join(metadata[genomeId]['taxonomy']) + '\t%.2f' % (float(metadata[genomeId]['genome size']) / 1e6) + '\t' + str(metadata[genomeId]['scaffold count']))
fout.write('\t' + metadata[genomeId]['biotic relationships'] + '\t' + metadata[genomeId]['status'])
for d in data:
fout.write('\t' + d[0] + '\t' + str(d[1]) + '\t' + str(d[2]) + '\t%.3f\t%.3f' % (d[3], d[4]))
fout.write('\n')
fout.close()
def run(self, ubiquityThreshold, singleCopyThreshold, minGenomes, minMarkers, mostSpecificRank, percentGenomes, numReplicates):
img = IMG()
lineages = img.lineagesByCriteria(minGenomes, mostSpecificRank)
fout = open('./data/lineage_evaluation.tsv', 'w')
fout.write('Lineage\t# genomes\t# markers\tpercentage\tnum replicates\tmean\tstd\tmean %\tmean + std%\tmean + 2*std %\n')
for lineage in lineages:
genomeIds = img.genomeIdsByTaxonomy(lineage, 'Final')
if len(genomeIds) < minGenomes:
continue
countTable = img.countTable(genomeIds)
countTable = img.filterTable(genomeIds, countTable, ubiquityThreshold*0.9, singleCopyThreshold*0.9)
# calculate marker set for all genomes
markerGenes = img.markerGenes(genomeIds, countTable, ubiquityThreshold*len(genomeIds), singleCopyThreshold*len(genomeIds))
if len(markerGenes) < minMarkers:
continue
print '\nLineage ' + lineage + ' contains ' + str(len(genomeIds)) + ' genomes.'
print ' Marker genes: ' + str(len(markerGenes))
fout.write(lineage + '\t' + str(len(genomeIds)) + '\t' + str(len(markerGenes)) + '\t%.2f' % percentGenomes + '\t' + str(numReplicates))
# withhold select percentage of genomes and calculate new marker set
changeMarkerSetSize = []
for _ in xrange(0, numReplicates):
subsetGenomeIds = random.sample(genomeIds, int((1.0-percentGenomes)*len(genomeIds) + 0.5))
newMarkerGenes = img.markerGenes(subsetGenomeIds, countTable, ubiquityThreshold*len(subsetGenomeIds), singleCopyThreshold*len(subsetGenomeIds))
changeMarkerSetSize.append(len(newMarkerGenes.symmetric_difference(markerGenes)))
m = mean(changeMarkerSetSize)
s = std(changeMarkerSetSize)
print ' Mean: %.2f, Std: %.2f, Per: %.2f' % (m, s, (m+ 2*s) * 100 / len(markerGenes))
fout.write('\t%.2f\t%.2f\t%.2f\t%.2f\t%.2f' % (m, s, m * 100 / len(markerGenes), (m + s) * 100 / len(markerGenes), (m + 2*s) * 100 / len(markerGenes)) + '\n')
fout.close()
def run(self, ubiquityThreshold, singleCopyThreshold, minGenomes,
mostSpecificRank, minMarkers, completenessThreshold,
contaminationThreshold):
print 'Ubiquity threshold: ' + str(ubiquityThreshold)
print 'Single-copy threshold: ' + str(singleCopyThreshold)
print 'Min. genomes: ' + str(minGenomes)
print 'Most specific taxonomic rank: ' + str(mostSpecificRank)
print 'Min markers: ' + str(minMarkers)
print 'Completeness threshold: ' + str(completenessThreshold)
print 'Contamination threshold: ' + str(contaminationThreshold)
img = IMG()
markerset = MarkerSet()
lineages = img.lineagesByCriteria(minGenomes, mostSpecificRank)
degenerateGenomes = {}
for lineage in lineages:
genomeIds = img.genomeIdsByTaxonomy(lineage, 'Final')
print ''
print 'Lineage ' + lineage + ' contains ' + str(
len(genomeIds)) + ' genomes.'
# get table of PFAMs and do some initial filtering to remove PFAMs that are
# clearly not going to pass the ubiquity and single-copy thresholds
countTable = img.countTable(genomeIds)
countTable = img.filterTable(genomeIds, countTable,
ubiquityThreshold * 0.9,
singleCopyThreshold * 0.9)
markerGenes = markerset.markerGenes(
genomeIds, countTable, ubiquityThreshold * len(genomeIds),
singleCopyThreshold * len(genomeIds))
if len(markerGenes) < minMarkers:
continue
geneDistTable = img.geneDistTable(genomeIds,
markerGenes,
spacingBetweenContigs=1e6)
colocatedGenes = markerset.colocatedGenes(geneDistTable)
colocatedSets = markerset.colocatedSets(colocatedGenes,
markerGenes)
for genomeId in genomeIds:
completeness, contamination = markerset.genomeCheck(
colocatedSets, genomeId, countTable)
if completeness < completenessThreshold or contamination > contaminationThreshold:
degenerateGenomes[genomeId] = degenerateGenomes.get(
genomeId, []) + [[
lineage.split(';')[-1].strip(),
len(genomeIds),
len(colocatedSets), completeness, contamination
]]
# write out degenerate genomes
metadata = img.genomeMetadata('Final')
fout = open('./data/degenerate_genomes.tsv', 'w')
fout.write(
'Genome Id\tTaxonomy\tGenome Size (Gbps)\tScaffolds\tBiotic Relationships\tStatus\tLineage\t# genomes\tMarker set size\tCompleteness\tContamination\n'
)
for genomeId, data in degenerateGenomes.iteritems():
fout.write(genomeId + '\t' +
'; '.join(metadata[genomeId]['taxonomy']) + '\t%.2f' %
(float(metadata[genomeId]['genome size']) / 1e6) +
'\t' + str(metadata[genomeId]['scaffold count']))
fout.write('\t' + metadata[genomeId]['biotic relationships'] +
'\t' + metadata[genomeId]['status'])
for d in data:
fout.write('\t' + d[0] + '\t' + str(d[1]) + '\t' + str(d[2]) +
'\t%.3f\t%.3f' % (d[3], d[4]))
fout.write('\n')
fout.close()
def run(self, taxonomyStr, ubiquityThreshold, singleCopyThreshold, numBins,
numRndGenomes):
img = IMG()
markerSet = MarkerSet()
metadata = img.genomeMetadata()
lineageGenomeIds = img.genomeIdsByTaxonomy(taxonomyStr, metadata)
# build marker set from finished prokaryotic genomes
genomeIds = []
for genomeId in lineageGenomeIds:
if metadata[genomeId]['status'] == 'Finished' and (
metadata[genomeId]['taxonomy'][0] == 'Bacteria'
or metadata[genomeId]['taxonomy'][0] == 'Archaea'):
genomeIds.append(genomeId)
genomeIds = set(genomeIds) - img.genomesWithMissingData(genomeIds)
print 'Lineage ' + taxonomyStr + ' contains ' + str(
len(genomeIds)) + ' genomes.'
# get marker set
countTable = img.countTable(genomeIds)
countTable = img.filterTable(genomeIds, countTable,
0.9 * ubiquityThreshold,
0.9 * singleCopyThreshold)
markerGenes = markerSet.markerGenes(
genomeIds, countTable, ubiquityThreshold * len(genomeIds),
singleCopyThreshold * len(genomeIds))
tigrToRemove = img.identifyRedundantTIGRFAMs(markerGenes)
markerGenes = markerGenes - tigrToRemove
geneDistTable = img.geneDistTable(genomeIds,
markerGenes,
spacingBetweenContigs=1e6)
print 'Number of marker genes: ' + str(len(markerGenes))
# randomly set genomes to plot
if numRndGenomes != -1:
genomeIds = random.sample(list(genomeIds), numRndGenomes)
genomeIds = set(genomeIds)
# plot distribution of marker genes
filename = 'geneDistribution.' + taxonomyStr.replace(
';', '_') + '.' + str(ubiquityThreshold) + '-' + str(
singleCopyThreshold) + '.tsv'
fout = open(filename, 'w')
fout.write(
'Genome ID\tLineage\tNumber of Genes\tUniformity\tDistribution\n')
matrix = []
rowLabels = []
for genomeId in genomeIds:
binSize = float(metadata[genomeId]['genome size']) / numBins
binCounts = [0] * numBins
pts = []
for _, data in geneDistTable[genomeId].iteritems():
for genePos in data:
binNum = int(genePos[1] / binSize)
binCounts[binNum] += 1
pts.append(genePos[1])
matrix.append(binCounts)
u = markerSet.uniformity(metadata[genomeId]['genome size'], pts)
fout.write(genomeId + '\t' +
'; '.join(metadata[genomeId]['taxonomy']) + '\t' +
str(len(geneDistTable[genomeId])) + '\t%.3f' % u)
for b in xrange(0, numBins):
fout.write('\t' + str(binCounts[b]))
fout.write('\n')
rowLabels.append('%.2f' % u + ', ' + str(genomeId) + ' - ' +
'; '.join(metadata[genomeId]['taxonomy'][0:5]))
fout.close()
# plot data
heatmap = Heatmap()
plotFilename = 'geneDistribution.' + taxonomyStr.replace(
';', '_') + '.' + str(ubiquityThreshold) + '-' + str(
singleCopyThreshold) + '.png'
heatmap.plot(plotFilename, matrix, rowLabels, 0.6)
def run(self, ubiquityThreshold, singleCopyThreshold, minGenomes,
minMarkers, mostSpecificRank, percentGenomes, numReplicates):
img = IMG()
lineages = img.lineagesByCriteria(minGenomes, mostSpecificRank)
fout = open('./data/lineage_evaluation.tsv', 'w')
fout.write(
'Lineage\t# genomes\t# markers\tpercentage\tnum replicates\tmean\tstd\tmean %\tmean + std%\tmean + 2*std %\n'
)
for lineage in lineages:
genomeIds = img.genomeIdsByTaxonomy(lineage, 'Final')
if len(genomeIds) < minGenomes:
continue
countTable = img.countTable(genomeIds)
countTable = img.filterTable(genomeIds, countTable,
ubiquityThreshold * 0.9,
singleCopyThreshold * 0.9)
# calculate marker set for all genomes
markerGenes = img.markerGenes(genomeIds, countTable,
ubiquityThreshold * len(genomeIds),
singleCopyThreshold * len(genomeIds))
if len(markerGenes) < minMarkers:
continue
print '\nLineage ' + lineage + ' contains ' + str(
len(genomeIds)) + ' genomes.'
print ' Marker genes: ' + str(len(markerGenes))
fout.write(lineage + '\t' + str(len(genomeIds)) + '\t' +
str(len(markerGenes)) + '\t%.2f' % percentGenomes +
'\t' + str(numReplicates))
# withhold select percentage of genomes and calculate new marker set
changeMarkerSetSize = []
for _ in xrange(0, numReplicates):
subsetGenomeIds = random.sample(
genomeIds,
int((1.0 - percentGenomes) * len(genomeIds) + 0.5))
newMarkerGenes = img.markerGenes(
subsetGenomeIds, countTable,
ubiquityThreshold * len(subsetGenomeIds),
singleCopyThreshold * len(subsetGenomeIds))
changeMarkerSetSize.append(
len(newMarkerGenes.symmetric_difference(markerGenes)))
m = mean(changeMarkerSetSize)
s = std(changeMarkerSetSize)
print ' Mean: %.2f, Std: %.2f, Per: %.2f' % (m, s,
(m + 2 * s) * 100 /
len(markerGenes))
fout.write('\t%.2f\t%.2f\t%.2f\t%.2f\t%.2f' %
(m, s, m * 100 / len(markerGenes),
(m + s) * 100 / len(markerGenes),
(m + 2 * s) * 100 / len(markerGenes)) + '\n')
fout.close()
def run(self, taxonomyStr, ubiquityThreshold, singleCopyThreshold, numBins, numRndGenomes):
img = IMG()
markerSet = MarkerSet()
metadata = img.genomeMetadata()
lineageGenomeIds = img.genomeIdsByTaxonomy(taxonomyStr, metadata)
# build marker set from finished prokaryotic genomes
genomeIds = []
for genomeId in lineageGenomeIds:
if metadata[genomeId]['status'] == 'Finished' and (metadata[genomeId]['taxonomy'][0] == 'Bacteria' or metadata[genomeId]['taxonomy'][0] == 'Archaea'):
genomeIds.append(genomeId)
genomeIds = set(genomeIds) - img.genomesWithMissingData(genomeIds)
print 'Lineage ' + taxonomyStr + ' contains ' + str(len(genomeIds)) + ' genomes.'
# get marker set
countTable = img.countTable(genomeIds)
countTable = img.filterTable(genomeIds, countTable, 0.9*ubiquityThreshold, 0.9*singleCopyThreshold)
markerGenes = markerSet.markerGenes(genomeIds, countTable, ubiquityThreshold*len(genomeIds), singleCopyThreshold*len(genomeIds))
tigrToRemove = img.identifyRedundantTIGRFAMs(markerGenes)
markerGenes = markerGenes - tigrToRemove
geneDistTable = img.geneDistTable(genomeIds, markerGenes, spacingBetweenContigs=1e6)
print 'Number of marker genes: ' + str(len(markerGenes))
# randomly set genomes to plot
if numRndGenomes != -1:
genomeIds = random.sample(list(genomeIds), numRndGenomes)
genomeIds = set(genomeIds)
# plot distribution of marker genes
filename = 'geneDistribution.' + taxonomyStr.replace(';','_') + '.' + str(ubiquityThreshold) + '-' + str(singleCopyThreshold) + '.tsv'
fout = open(filename, 'w')
fout.write('Genome ID\tLineage\tNumber of Genes\tUniformity\tDistribution\n')
matrix = []
rowLabels = []
for genomeId in genomeIds:
binSize = float(metadata[genomeId]['genome size']) / numBins
binCounts = [0]*numBins
pts = []
for _, data in geneDistTable[genomeId].iteritems():
for genePos in data:
binNum = int(genePos[1] / binSize)
binCounts[binNum] += 1
pts.append(genePos[1])
matrix.append(binCounts)
u = markerSet.uniformity(metadata[genomeId]['genome size'], pts)
fout.write(genomeId + '\t' + '; '.join(metadata[genomeId]['taxonomy']) + '\t' + str(len(geneDistTable[genomeId])) + '\t%.3f' % u)
for b in xrange(0, numBins):
fout.write('\t' + str(binCounts[b]))
fout.write('\n')
rowLabels.append('%.2f' % u + ', ' + str(genomeId) + ' - ' + '; '.join(metadata[genomeId]['taxonomy'][0:5]))
fout.close()
# plot data
heatmap = Heatmap()
plotFilename = 'geneDistribution.' + taxonomyStr.replace(';','_') + '.' + str(ubiquityThreshold) + '-' + str(singleCopyThreshold) + '.png'
heatmap.plot(plotFilename, matrix, rowLabels, 0.6)
def run(self, ubiquityThreshold, singleCopyThreshold, trustedCompleteness, trustedContamination, genomeCompleteness, genomeContamination):
img = IMG()
markerset = MarkerSet()
metadata = img.genomeMetadata()
trustedOut = open('./data/trusted_genomes.tsv', 'w')
trustedOut.write('Genome Id\tLineage\tGenome size (Mbps)\tScaffold count\tBiotic Relationship\tStatus\tCompleteness\tContamination\n')
filteredOut = open('./data/filtered_genomes.tsv', 'w')
filteredOut.write('Genome Id\tLineage\tGenome size (Mbps)\tScaffold count\tBiotic Relationship\tStatus\tCompleteness\tContamination\n')
allGenomeIds = set()
allTrustedGenomeIds = set()
for lineage in ['Archaea', 'Bacteria']:
# get all genomes in lineage and build gene count table
print '\nBuilding gene count table.'
allLineageGenomeIds = img.genomeIdsByTaxonomy(lineage, metadata, 'All')
countTable = img.countTable(allLineageGenomeIds)
countTable = img.filterTable(allLineageGenomeIds, countTable, 0.9*ubiquityThreshold, 0.9*singleCopyThreshold)
# get all genomes from specific lineage
allGenomeIds = allGenomeIds.union(allLineageGenomeIds)
print 'Lineage ' + lineage + ' contains ' + str(len(allLineageGenomeIds)) + ' genomes.'
# tabulate genomes from each phylum
allPhylumCounts = {}
for genomeId in allLineageGenomeIds:
taxon = metadata[genomeId]['taxonomy'][1]
allPhylumCounts[taxon] = allPhylumCounts.get(taxon, 0) + 1
# identify marker set for genomes
markerGenes = markerset.markerGenes(allLineageGenomeIds, countTable, ubiquityThreshold*len(allLineageGenomeIds), singleCopyThreshold*len(allLineageGenomeIds))
print ' Marker genes: ' + str(len(markerGenes))
geneDistTable = img.geneDistTable(allLineageGenomeIds, markerGenes, spacingBetweenContigs=1e6)
colocatedGenes = markerset.colocatedGenes(geneDistTable, metadata)
colocatedSets = markerset.colocatedSets(colocatedGenes, markerGenes)
print ' Marker set size: ' + str(len(colocatedSets))
# identifying trusted genomes (highly complete, low contamination genomes)
trustedGenomeIds = set()
for genomeId in allLineageGenomeIds:
completeness, contamination = markerset.genomeCheck(colocatedSets, genomeId, countTable)
if completeness >= trustedCompleteness and contamination <= trustedContamination:
trustedGenomeIds.add(genomeId)
allTrustedGenomeIds.add(genomeId)
trustedOut.write(genomeId + '\t' + '; '.join(metadata[genomeId]['taxonomy']))
trustedOut.write('\t%.2f' % (float(metadata[genomeId]['genome size']) / 1e6))
trustedOut.write('\t' + str(metadata[genomeId]['scaffold count']))
trustedOut.write('\t' + metadata[genomeId]['biotic relationships'])
trustedOut.write('\t' + metadata[genomeId]['status'])
trustedOut.write('\t%.3f\t%.3f' % (completeness, contamination) + '\n')
else:
filteredOut.write(genomeId + '\t' + '; '.join(metadata[genomeId]['taxonomy']))
filteredOut.write('\t%.2f' % (float(metadata[genomeId]['genome size']) / 1e6))
filteredOut.write('\t' + str(metadata[genomeId]['scaffold count']))
filteredOut.write('\t' + metadata[genomeId]['biotic relationships'])
filteredOut.write('\t' + metadata[genomeId]['status'])
filteredOut.write('\t%.3f\t%.3f' % (completeness, contamination) + '\n')
print ' Trusted genomes: ' + str(len(trustedGenomeIds))
# determine status of trusted genomes
statusBreakdown = {}
for genomeId in trustedGenomeIds:
statusBreakdown[metadata[genomeId]['status']] = statusBreakdown.get(metadata[genomeId]['status'], 0) + 1
print ' Trusted genome status breakdown: '
for status, count in statusBreakdown.iteritems():
print ' ' + status + ': ' + str(count)
# determine status of retained genomes
proposalNameBreakdown = {}
for genomeId in trustedGenomeIds:
proposalNameBreakdown[metadata[genomeId]['proposal name']] = proposalNameBreakdown.get(metadata[genomeId]['proposal name'], 0) + 1
print ' Retained genome proposal name breakdown: '
for pn, count in proposalNameBreakdown.iteritems():
if 'KMG' in pn or 'GEBA' in pn or 'HMP' in pn:
print ' ' + pn + ': ' + str(count)
print ' Filtered genomes by phylum:'
trustedPhylumCounts = {}
for genomeId in trustedGenomeIds:
taxon = metadata[genomeId]['taxonomy'][1]
trustedPhylumCounts[taxon] = trustedPhylumCounts.get(taxon, 0) + 1
for phylum, count in allPhylumCounts.iteritems():
print phylum + ': %d of %d' % (trustedPhylumCounts.get(phylum, 0), count)
trustedOut.close()
filteredOut.close()
# write out lineage statistics for genome distribution
allStats = {}
trustedStats = {}
for r in xrange(0, 6): # Domain to Genus
for genomeId, data in metadata.iteritems():
taxaStr = '; '.join(data['taxonomy'][0:r+1])
allStats[taxaStr] = allStats.get(taxaStr, 0) + 1
if genomeId in allTrustedGenomeIds:
trustedStats[taxaStr] = trustedStats.get(taxaStr, 0) + 1
sortedLineages = img.lineagesSorted()
fout = open('./data/lineage_stats.tsv', 'w')
fout.write('Lineage\tGenomes with metadata\tTrusted genomes\n')
for lineage in sortedLineages:
fout.write(lineage + '\t' + str(allStats.get(lineage, 0))+ '\t' + str(trustedStats.get(lineage, 0))+ '\n')
fout.close()
def run(self, ubiquityThreshold, singleCopyThreshold, trustedCompleteness,
trustedContamination, genomeCompleteness, genomeContamination):
img = IMG()
markerset = MarkerSet()
metadata = img.genomeMetadata()
trustedOut = open('./data/trusted_genomes.tsv', 'w')
trustedOut.write(
'Genome Id\tLineage\tGenome size (Mbps)\tScaffold count\tBiotic Relationship\tStatus\tCompleteness\tContamination\n'
)
filteredOut = open('./data/filtered_genomes.tsv', 'w')
filteredOut.write(
'Genome Id\tLineage\tGenome size (Mbps)\tScaffold count\tBiotic Relationship\tStatus\tCompleteness\tContamination\n'
)
allGenomeIds = set()
allTrustedGenomeIds = set()
for lineage in ['Archaea', 'Bacteria']:
# get all genomes in lineage and build gene count table
print '\nBuilding gene count table.'
allLineageGenomeIds = img.genomeIdsByTaxonomy(
lineage, metadata, 'All')
countTable = img.countTable(allLineageGenomeIds)
countTable = img.filterTable(allLineageGenomeIds, countTable,
0.9 * ubiquityThreshold,
0.9 * singleCopyThreshold)
# get all genomes from specific lineage
allGenomeIds = allGenomeIds.union(allLineageGenomeIds)
print 'Lineage ' + lineage + ' contains ' + str(
len(allLineageGenomeIds)) + ' genomes.'
# tabulate genomes from each phylum
allPhylumCounts = {}
for genomeId in allLineageGenomeIds:
taxon = metadata[genomeId]['taxonomy'][1]
allPhylumCounts[taxon] = allPhylumCounts.get(taxon, 0) + 1
# identify marker set for genomes
markerGenes = markerset.markerGenes(
allLineageGenomeIds, countTable,
ubiquityThreshold * len(allLineageGenomeIds),
singleCopyThreshold * len(allLineageGenomeIds))
print ' Marker genes: ' + str(len(markerGenes))
geneDistTable = img.geneDistTable(allLineageGenomeIds,
markerGenes,
spacingBetweenContigs=1e6)
colocatedGenes = markerset.colocatedGenes(geneDistTable, metadata)
colocatedSets = markerset.colocatedSets(colocatedGenes,
markerGenes)
print ' Marker set size: ' + str(len(colocatedSets))
# identifying trusted genomes (highly complete, low contamination genomes)
trustedGenomeIds = set()
for genomeId in allLineageGenomeIds:
completeness, contamination = markerset.genomeCheck(
colocatedSets, genomeId, countTable)
if completeness >= trustedCompleteness and contamination <= trustedContamination:
trustedGenomeIds.add(genomeId)
allTrustedGenomeIds.add(genomeId)
trustedOut.write(genomeId + '\t' +
'; '.join(metadata[genomeId]['taxonomy']))
trustedOut.write(
'\t%.2f' %
(float(metadata[genomeId]['genome size']) / 1e6))
trustedOut.write('\t' +
str(metadata[genomeId]['scaffold count']))
trustedOut.write(
'\t' + metadata[genomeId]['biotic relationships'])
trustedOut.write('\t' + metadata[genomeId]['status'])
trustedOut.write('\t%.3f\t%.3f' %
(completeness, contamination) + '\n')
else:
filteredOut.write(
genomeId + '\t' +
'; '.join(metadata[genomeId]['taxonomy']))
filteredOut.write(
'\t%.2f' %
(float(metadata[genomeId]['genome size']) / 1e6))
filteredOut.write(
'\t' + str(metadata[genomeId]['scaffold count']))
filteredOut.write(
'\t' + metadata[genomeId]['biotic relationships'])
filteredOut.write('\t' + metadata[genomeId]['status'])
filteredOut.write('\t%.3f\t%.3f' %
(completeness, contamination) + '\n')
print ' Trusted genomes: ' + str(len(trustedGenomeIds))
# determine status of trusted genomes
statusBreakdown = {}
for genomeId in trustedGenomeIds:
statusBreakdown[metadata[genomeId]
['status']] = statusBreakdown.get(
metadata[genomeId]['status'], 0) + 1
print ' Trusted genome status breakdown: '
for status, count in statusBreakdown.iteritems():
print ' ' + status + ': ' + str(count)
# determine status of retained genomes
proposalNameBreakdown = {}
for genomeId in trustedGenomeIds:
proposalNameBreakdown[metadata[genomeId][
'proposal name']] = proposalNameBreakdown.get(
metadata[genomeId]['proposal name'], 0) + 1
print ' Retained genome proposal name breakdown: '
for pn, count in proposalNameBreakdown.iteritems():
if 'KMG' in pn or 'GEBA' in pn or 'HMP' in pn:
print ' ' + pn + ': ' + str(count)
print ' Filtered genomes by phylum:'
trustedPhylumCounts = {}
for genomeId in trustedGenomeIds:
taxon = metadata[genomeId]['taxonomy'][1]
trustedPhylumCounts[taxon] = trustedPhylumCounts.get(taxon,
0) + 1
for phylum, count in allPhylumCounts.iteritems():
print phylum + ': %d of %d' % (trustedPhylumCounts.get(
phylum, 0), count)
trustedOut.close()
filteredOut.close()
# write out lineage statistics for genome distribution
allStats = {}
trustedStats = {}
for r in xrange(0, 6): # Domain to Genus
for genomeId, data in metadata.iteritems():
taxaStr = '; '.join(data['taxonomy'][0:r + 1])
allStats[taxaStr] = allStats.get(taxaStr, 0) + 1
if genomeId in allTrustedGenomeIds:
trustedStats[taxaStr] = trustedStats.get(taxaStr, 0) + 1
sortedLineages = img.lineagesSorted()
fout = open('./data/lineage_stats.tsv', 'w')
fout.write('Lineage\tGenomes with metadata\tTrusted genomes\n')
for lineage in sortedLineages:
fout.write(lineage + '\t' + str(allStats.get(lineage, 0)) + '\t' +
str(trustedStats.get(lineage, 0)) + '\n')
fout.close()
def run(self, taxonomyStr, ubiquityThreshold, singleCopyThreshold,
replicates, minGenomes, maxGenomes, stepSize):
img = IMG()
markergenes = MarkerGenes()
genomeIds = img.genomeIdsByTaxonomy(taxonomyStr, 'Final')
print('Lineage ' + taxonomyStr + ' contains ' + str(len(genomeIds)) +
' genomes.')
if len(genomeIds) < minGenomes:
sys.stderr.write('[Error] Insufficent number of genomes.\n')
sys.exit()
print('')
print('Ubiquity threshold: ' + str(ubiquityThreshold))
print('Single-copy threshold: ' + str(singleCopyThreshold))
meanMarkerSetSize = []
stdMarkerSetSize = []
markerSetSizes = []
if maxGenomes == -1:
maxGenomes = len(genomeIds)
if maxGenomes > len(genomeIds):
maxGenomes = len(genomeIds)
countTable = img.countTable(genomeIds)
countTable = img.filterTable(genomeIds, countTable)
for numGenomes in range(minGenomes, maxGenomes, stepSize):
markerSetSize = []
for _ in range(0, replicates):
genomeIdSubset = random.sample(genomeIds, numGenomes)
markerGenes = markergenes.identify(
genomeIdSubset, countTable,
ubiquityThreshold * len(genomeIdSubset),
singleCopyThreshold * len(genomeIdSubset))
geneDistTable = img.geneDistTable(genomeIdSubset,
markerGenes,
spacingBetweenContigs=1e6)
colocatedGenes = img.colocatedGenes(geneDistTable)
colocatedSets = img.colocatedSets(colocatedGenes, markerGenes)
markerSetSize.append(len(colocatedSets))
markerSetSizes.append(markerSetSize)
m = mean(markerSetSize)
meanMarkerSetSize.append(m)
s = std(markerSetSize)
stdMarkerSetSize.append(s)
print('')
print('Genomes: ' + str(numGenomes) + ', Ubiquity > ' +
str(int(ubiquityThreshold * len(genomeIdSubset))) +
', Single-copy > ' +
str(int(singleCopyThreshold * len(genomeIdSubset))))
print('Mean: %.2f +/- %.2f' % (m, s))
print('Min: %d, Max: %d' %
(min(markerSetSize), max(markerSetSize)))
# plot data
errorBar = ErrorBar()
plotFilename = './images/markerset.' + taxonomyStr.replace(
';', '_') + '.' + str(ubiquityThreshold) + '-' + str(
singleCopyThreshold) + '.errorbar.png'
title = taxonomyStr.replace(
';', '; '
) + '\n' + 'Ubiquity = %.2f' % ubiquityThreshold + ', Single-copy = %.2f' % singleCopyThreshold
errorBar.plot(plotFilename, arange(minGenomes, maxGenomes, stepSize),
meanMarkerSetSize, stdMarkerSetSize, 'Number of Genomes',
'Marker Set Size', title)
boxPlot = BoxPlot()
plotFilename = './images/markerset.' + taxonomyStr.replace(
';', '_') + '.' + str(ubiquityThreshold) + '-' + str(
singleCopyThreshold) + '.boxplot.png'
boxPlot.plot(plotFilename, markerSetSizes,
arange(minGenomes, maxGenomes, stepSize),
'Number of Genomes', 'Marker Set Size', True, title)
def run(self, taxonomyStr, ubiquityThreshold, singleCopyThreshold, replicates, minGenomes, maxGenomes, stepSize):
img = IMG()
markergenes = MarkerGenes()
genomeIds = img.genomeIdsByTaxonomy(taxonomyStr, 'Final')
print 'Lineage ' + taxonomyStr + ' contains ' + str(len(genomeIds)) + ' genomes.'
if len(genomeIds) < minGenomes:
sys.stderr.write('[Error] Insufficent number of genomes.\n')
sys.exit()
print ''
print 'Ubiquity threshold: ' + str(ubiquityThreshold)
print 'Single-copy threshold: ' + str(singleCopyThreshold)
meanMarkerSetSize = []
stdMarkerSetSize = []
markerSetSizes = []
if maxGenomes == -1:
maxGenomes = len(genomeIds)
if maxGenomes > len(genomeIds):
maxGenomes = len(genomeIds)
countTable = img.countTable(genomeIds)
countTable = img.filterTable(genomeIds, countTable)
for numGenomes in xrange(minGenomes, maxGenomes, stepSize):
markerSetSize = []
for _ in xrange(0, replicates):
genomeIdSubset = random.sample(genomeIds, numGenomes)
markerGenes = markergenes.identify(genomeIdSubset, countTable, ubiquityThreshold*len(genomeIdSubset), singleCopyThreshold*len(genomeIdSubset))
geneDistTable = img.geneDistTable(genomeIdSubset, markerGenes, spacingBetweenContigs=1e6)
colocatedGenes = img.colocatedGenes(geneDistTable)
colocatedSets = img.colocatedSets(colocatedGenes, markerGenes)
markerSetSize.append(len(colocatedSets))
markerSetSizes.append(markerSetSize)
m = mean(markerSetSize)
meanMarkerSetSize.append(m)
s = std(markerSetSize)
stdMarkerSetSize.append(s)
print ''
print 'Genomes: ' + str(numGenomes) + ', Ubiquity > ' + str(int(ubiquityThreshold*len(genomeIdSubset))) + ', Single-copy > ' + str(int(singleCopyThreshold*len(genomeIdSubset)))
print 'Mean: %.2f +/- %.2f' % (m, s)
print 'Min: %d, Max: %d' %(min(markerSetSize), max(markerSetSize))
# plot data
errorBar = ErrorBar()
plotFilename = './images/markerset.' + taxonomyStr.replace(';','_') + '.' + str(ubiquityThreshold) + '-' + str(singleCopyThreshold) + '.errorbar.png'
title = taxonomyStr.replace(';', '; ') + '\n' + 'Ubiquity = %.2f' % ubiquityThreshold + ', Single-copy = %.2f' % singleCopyThreshold
errorBar.plot(plotFilename, arange(minGenomes, maxGenomes, stepSize), meanMarkerSetSize, stdMarkerSetSize, 'Number of Genomes', 'Marker Set Size', title)
boxPlot = BoxPlot()
plotFilename = './images/markerset.' + taxonomyStr.replace(';','_') + '.' + str(ubiquityThreshold) + '-' + str(singleCopyThreshold) + '.boxplot.png'
boxPlot.plot(plotFilename, markerSetSizes, arange(minGenomes, maxGenomes, stepSize), 'Number of Genomes', 'Marker Set Size', True, title)
