def doit():
rug = Taxonomy.getTaxonomy('scratch/Ruggiero/', 'rug')
ott = Taxonomy.getTaxonomy('tax/ott/', 'ott')
union = UnionTaxonomy.newTaxonomy('ott')
union.absorb(rug)
union.absorb(ott)
union.dump('scratch/compare_Ruggiero/', '\t')
def combine(t, s, bluster):
u = UnionTaxonomy.newTaxonomy('union')
u.blustery = 0
ta = u.alignment(t)
u.align(ta)
u.merge(ta)
u.blustery = bluster
sa = u.alignment(s)
for root in s.roots():
sa.alignTaxon(root)
u.align(sa)
u.merge(sa)
u.check()
return u
def combine(t, s, bluster):
u = UnionTaxonomy.newTaxonomy('union')
u.blustery = 0
ta = u.alignment(t)
u.align(ta)
u.merge(ta)
u.blustery = bluster
sa = u.alignment(s)
for root in s.roots():
sa.alignTaxon(root)
u.align(sa)
u.merge(sa)
u.check()
return u
def combine(sep, t, s, bluster):
u = UnionTaxonomy.newTaxonomy('union')
u.blustery = 0
u.setSkeleton(sep)
ta = u.alignment(t)
u.align(ta)
u.merge(ta)
print u.lookup('a')
u.blustery = bluster
print s.lookup('a')
sa = u.alignment(s)
for root in s.roots():
sa.alignTaxon(root)
u.align(sa)
debug_alignment(sa)
return u
def create_ott(ott_spec):
# Fail fast
additions_clone_path = os.path.join(access_head('amendments'),
'amendments-1')
if not os.path.isdir(additions_clone_path):
print '# cannot find', additions_clone_path
sys.exit(1)
with open(os.path.join(access_head('idlist'), 'by_qid.csv'),
'r') as infile:
print '# can access idlist'
ott_path = management.source_path(ott_spec)
ott = UnionTaxonomy.newTaxonomy('ott')
# Would be nice if there were tests for all of these...
for name in names_of_interest:
ott.eventLogger.namesOfInterest.add(name)
ott.setSkeleton(Taxonomy.getTaxonomy('curation/separation/', 'separation'))
# These are particularly hard cases; create alignment targets up front
adjustments.deal_with_polysemies(ott)
# Align and merge each source in sequence
merge_sources(ott)
# "Old" patch system
TsvEdits.edit(ott, 'curation/edits/')
# consider try: ... except: print '**** Exception in patch_ott'
amendments.patch_ott(ott)
# End of topology changes. Now assign ids.
retain_ids(ott, access_source('ott-PREVIOUS'),
os.path.join(access_head('idlist'), 'by_qid.csv'))
# Apply the additions (which already have ids assigned).
# This has to happen *after* ids are assigned, since additions use OTT
# ids to identify parents.
print '-- Processing additions --'
Addition.processAdditions(additions_clone_path, ott)
# Mint ids for new nodes
print '-- Minting new ids --'
ott.assignNewIds(new_taxa_path)
# Remove all trees but the largest (or make them life incertae sedis)
ott.deforestate()
# data structure integrity checks
ott.check()
# For deprecated id report (dump)
ott.loadPreferredIds('ids_that_are_otus.tsv', False)
ott.loadPreferredIds('ids_in_synthesis.tsv', True)
ott.dump(ott_path)
record_ott_sources(ott_spec)
return ott
import sys, os, csv
from org.opentreeoflife.taxa import Taxonomy, SourceTaxonomy, Taxon
from org.opentreeoflife.smasher import UnionTaxonomy
dwh = UnionTaxonomy.newTaxonomy('dwh')
#Use this to tell smasher what separation file to use
dwh.setSkeleton(Taxonomy.getTaxonomy('tax/separation/', 'separation'))
# 1. trunk
# 2. ictv
# 3. IOC
# 4. ASW
# 5. ODO
# 6. BOM
# 7. ERE
# 8. ONY
# 9. EET
# 10. NCBI
# 11. WOR
# 12. CLP
# 13. COL
#use this to load the taxonomies
trunk = Taxonomy.getTaxonomy('t/tax/2018_12/trunk/', 'trunk')
ictv = Taxonomy.getTaxonomy('t/tax/2018_12/ictv/', 'ictv')
IOC = Taxonomy.getTaxonomy('t/tax/2018_12/IOC/', 'IOC')
ASW = Taxonomy.getTaxonomy('t/tax/2018_12/ASW/', 'ASW')
ODO = Taxonomy.getTaxonomy('t/tax/2018_12/ODO/', 'ODO')
# Command line argument = file to write to
# Writes a row for every OTT id that
# (a) occurs in tax/ott/,
# (b) occurs as an OTU in phylesystem,
# (c) is sourced only from in IRMNG.
import csv, sys
from org.opentreeoflife.taxa import Taxonomy, Rank
from org.opentreeoflife.smasher import UnionTaxonomy
union = UnionTaxonomy.newTaxonomy('ott')
union.loadPreferredIds('ids_that_are_otus.tsv', False)
union.loadPreferredIds('ids_in_synthesis.tsv', True)
ott = Taxonomy.getTaxonomy('tax/ott/', 'ott')
#ott = Taxonomy.getTaxonomy('t/tax/aster/', 'ott')
with open(sys.argv[1], 'w') as outfile:
writer = csv.writer(outfile)
writer.writerow(['irmng','ott','name','synthesis'])
for taxon in ott.taxa():
# if (taxon.rank == Rank.SPECIES_RANK and ...)
if (len(taxon.sourceIds) == 1 and
taxon.sourceIds[0].prefix == 'irmng'):
probe = union.importantIds.lookupId(taxon.id)
if probe != None:
writer.writerow([taxon.sourceIds[0].id,
taxon.id,
taxon.name,
'synthesis' if probe.inSynthesis else ''])
# Command line argument = file to write to
# Writes a row for every OTT id that
# (a) occurs in tax/ott/,
# (b) occurs as an OTU in phylesystem,
# (c) is sourced only from in IRMNG.
import csv, sys
from org.opentreeoflife.taxa import Taxonomy, Rank
from org.opentreeoflife.smasher import UnionTaxonomy
union = UnionTaxonomy.newTaxonomy('ott')
union.loadPreferredIds('ids_that_are_otus.tsv', False)
union.loadPreferredIds('ids_in_synthesis.tsv', True)
ott = Taxonomy.getTaxonomy('tax/ott/', 'ott')
#ott = Taxonomy.getTaxonomy('t/tax/aster/', 'ott')
with open(sys.argv[1], 'w') as outfile:
writer = csv.writer(outfile)
writer.writerow(['irmng', 'ott', 'name', 'synthesis'])
for taxon in ott.taxa():
# if (taxon.rank == Rank.SPECIES_RANK and ...)
if (len(taxon.sourceIds) == 1
and taxon.sourceIds[0].prefix == 'irmng'):
probe = union.importantIds.lookupId(taxon.id)
if probe != None:
writer.writerow([
taxon.sourceIds[0].id, taxon.id, taxon.name,
'synthesis' if probe.inSynthesis else ''
])
def create_ott():
ott = UnionTaxonomy.newTaxonomy()
# There ought to be tests for all of these...
for name in names_of_interest:
ott.eventlogger.namesOfInterest.add(name)
# When lumping, prefer to use ids that have been used in OTU matching
# This list could be used for all sorts of purposes...
ott.loadPreferredIds('ids-that-are-otus.tsv', False)
ott.loadPreferredIds('ids-in-synthesis.tsv', True)
ott.setSkeleton(Taxonomy.getTaxonomy('tax/skel/', 'skel'))
silva = prepare_silva(ott)
ott.absorb(silva)
check_invariants(ott)
h2007 = prepare_h2007(ott)
ott.absorb(h2007)
(fungi, fungorum_sans_fungi) = prepare_fungorum(ott)
ott.absorb(fungi)
check_invariants(ott)
# the non-Fungi from Index Fungorum get absorbed below
lamiales = prepare_lamiales(ott)
ott.absorb(lamiales)
(malacostraca, worms_sans_malacostraca) = prepare_worms(ott)
ott.absorb(malacostraca)
ncbi = prepare_ncbi(ott)
align_ncbi_to_silva(ncbi, silva, ott)
ott.absorb(ncbi)
check_invariants(ott)
ott.absorb(worms_sans_malacostraca)
ott.absorb(fungorum_sans_fungi)
gbif = prepare_gbif(ott)
ott.absorb(gbif)
irmng = prepare_irmng(ott)
ott.absorb(irmng)
taxonomies.link_to_h2007(ott)
get_default_extinct_info_from_gbif(gbif, ott)
check_invariants(ott)
# consider try: ... except: print '**** Exception in patch_ott'
patch_ott(ott)
# Experimental...
unextinct_ncbi(ncbi, ott)
# Remove all trees but the largest (or make them life incertae sedis)
ott.deforestate()
# -----------------------------------------------------------------------------
# OTT id assignment
# Force some id assignments... will try to automate this in the future.
# Most of these come from looking at the otu-deprecated.tsv file after a
# series of smasher runs.
for (inf, sup, id) in [
('Tipuloidea', 'Diptera', '722875'),
('Saccharomycetes', 'Saccharomycotina', '989999'),
('Phaeosphaeria', 'Ascomycota', '5486272'),
('Synedra acus','Eukaryota','992764'),
('Epiphloea','Halymeniaceae','5342325'),
('Hessea','Archaeplastida','600099'),
('Morganella','Arthropoda','6400'),
('Rhynchonelloidea','Rhynchonellidae','5316010'),
('Epiphloea', 'Lichinales', '5342482'),
('Morganella', 'Fungi', '973932'),
('Parmeliaceae', 'Lecanorales', '305904'),
]:
tax = ott.taxon(inf, sup)
if tax != None:
tax.setId(id)
ott.taxonThatContains('Rhynchonelloidea', 'Sphenarina').setId('795939') # NCBI
for (ncbi_id, ott_id, name) in ncbi_assignments_list:
n = ncbi.maybeTaxon(ncbi_id)
if n != None:
im = ott.image(n)
if im != None:
im.setId(ott_id)
else:
print '** NCBI %s not mapped - %s' % (ncbi_id, name)
else:
print '** No NCBI taxon %s - %s' % (ncbi_id, name)
# Cylindrocarpon is now Neonectria
ott.image(gbif.taxon('2563163')).setId('51754')
# Foo
trich = fungi.maybeTaxon('Trichosporon')
if trich != None:
ott.image(trich).setId('364222')
#ott.image(fungi.taxon('11060')).setId('4107132') #Cryptococcus - a total mess
# Assign OTT ids to taxa that don't have them, re-using old ids when possible
ids = Taxonomy.getTaxonomy('tax/prev_ott/')
# Assign old ids to nodes in the new version
ott.assignIds(ids)
report_on_h2007(h2007, ott)
return ott
def assemble():
# Create model taxonomy
tax = UnionTaxonomy.newTaxonomy('ott')
for name in ['Pentaphragma ellipticum',
'Lachnophyllum',
'Sipolisia',
'Cicerbita bourgaei',
'Adenophora triphylla',
'Artemisia vulgaris',
'Carlina libanotica',
]:
tax.watch(name)
# Establish homonym-resolution skeleton (not really used here)
# skel = Taxonomy.getTaxonomy('tax/skel/', 'skel')
# tax.setSkeleton(skel)
# Add NCBI subset to the model taxonomy
ncbi = Taxonomy.getTaxonomy('t/tax/ncbi_aster/', 'ncbi')
# analyzeOTUs sets flags on questionable taxa ("unclassified" and so on)
# to allow the option of suppression downstream
ncbi.analyzeOTUs()
align_and_merge(tax.alignment(ncbi))
# Add GBIF subset fo the model taxonomy
gbif = Taxonomy.getTaxonomy('t/tax/gbif_aster/', 'gbif')
gbif.smush()
# analyzeMajorRankConflicts sets the "major_rank_conflict" flag when
# intermediate ranks are missing (e.g. a family that's a child of a
# class)
gbif.analyzeMajorRankConflicts()
align_and_merge(tax.alignment(gbif))
# "Old" patch system with tab-delimited files
TsvEdits.edit(tax, 't/edits/')
props = [
has_parent(taxon('Phellinaceae'), taxon('Asterales'), 'test:1')
]
for prop in props:
print proclaim(tax, prop)
gen = tax.newTaxon("Opentreeia", "genus", "data:testing")
gen.take(tax.newTaxon("Opentreeia sp. C", "species", "data:testing"))
gen.take(tax.newTaxon("Opentreeia sp. D", "species", "data:testing"))
# Example of referring to a taxon
fam = tax.maybeTaxon("Phellinaceae")
if fam != None:
# Example of how you might add a genus to the taxonomy
fam.take(gen)
# Test deletion feature
sp = tax.newTaxon("Opentreeia sp. C", "species", "data:testing")
gen.take(sp)
sp.prune("aster.py")
# tax.loadPreferredIds('ids-that-are-otus.tsv')
additions_repo_path = 't/feed/amendments/amendments-0'
new_taxa_path = 't/new_taxa'
# Assign identifiers to the taxa in the model taxonomy. Identifiers
# assigned in the previous version are carried over to this version.
ids = Taxonomy.getTaxonomy('t/tax/prev_aster/', 'ott')
tax.carryOverIds(ids) # performs alignment
Addition.processAdditions(additions_repo_path, tax)
if False: # too slow for everyday testing purposes.
print '-- Checking id list'
assign_ids_from_list(tax, 'ott_id_list/by_qid.csv')
tax.assignNewIds(new_taxa_path)
tax.check()
# Write the model taxonomy out to a set of files
tax.dump('t/tax/aster/', '\t|\t')
def assemble():
# Create model taxonomy
tax = UnionTaxonomy.newTaxonomy('ott')
for name in [
'Pentaphragma ellipticum',
'Lachnophyllum',
'Sipolisia',
'Cicerbita bourgaei',
'Adenophora triphylla',
'Artemisia vulgaris',
'Carlina libanotica',
]:
tax.watch(name)
# Establish homonym-resolution skeleton (not really used here)
# skel = Taxonomy.getTaxonomy('tax/skel/', 'skel')
# tax.setSkeleton(skel)
# Add NCBI subset to the model taxonomy
ncbi = Taxonomy.getTaxonomy('t/tax/ncbi_aster/', 'ncbi')
# analyzeOTUs sets flags on questionable taxa ("unclassified" and so on)
# to allow the option of suppression downstream
ncbi.analyzeOTUs()
align_and_merge(tax.alignment(ncbi))
# Add GBIF subset fo the model taxonomy
gbif = Taxonomy.getTaxonomy('t/tax/gbif_aster/', 'gbif')
gbif.smush()
# analyzeMajorRankConflicts sets the "major_rank_conflict" flag when
# intermediate ranks are missing (e.g. a family that's a child of a
# class)
gbif.analyzeMajorRankConflicts()
align_and_merge(tax.alignment(gbif))
# "Old" patch system with tab-delimited files
TsvEdits.edit(tax, 't/edits/')
props = [has_parent(taxon('Phellinaceae'), taxon('Asterales'), 'test:1')]
for prop in props:
print proclaim(tax, prop)
gen = tax.newTaxon("Opentreeia", "genus", "data:testing")
gen.take(tax.newTaxon("Opentreeia sp. C", "species", "data:testing"))
gen.take(tax.newTaxon("Opentreeia sp. D", "species", "data:testing"))
# Example of referring to a taxon
fam = tax.maybeTaxon("Phellinaceae")
if fam != None:
# Example of how you might add a genus to the taxonomy
fam.take(gen)
# Test deletion feature
sp = tax.newTaxon("Opentreeia sp. C", "species", "data:testing")
gen.take(sp)
sp.prune("aster.py")
# tax.loadPreferredIds('ids-that-are-otus.tsv')
additions_repo_path = 't/feed/amendments/amendments-0'
new_taxa_path = 't/new_taxa'
# Assign identifiers to the taxa in the model taxonomy. Identifiers
# assigned in the previous version are carried over to this version.
ids = Taxonomy.getTaxonomy('t/tax/prev_aster/', 'ott')
tax.carryOverIds(ids) # performs alignment
Addition.processAdditions(additions_repo_path, tax)
if False: # too slow for everyday testing purposes.
print '-- Checking id list'
assign_ids_from_list(tax, 'ott_id_list/by_qid.csv')
tax.assignNewIds(new_taxa_path)
tax.check()
# Write the model taxonomy out to a set of files
tax.dump('t/tax/aster/', '\t|\t')