def test():
tax = Taxonomy.newTaxonomy()
tax.newTaxon('Mouse', 'species', 'about:blank')
tax.newTaxon('Dog', 'species', 'about:blank')
tax.newTaxon('Mammal', 'class', 'about:blank')
new_claims = [
Has_child('Mammal', 'Mouse', 'about:blank'),
Whether_same('Mouse', 'Mus', True),
Whether_same('Muus', 'Mouse', True),
]
expectations = [
Has_child('Mammal', 'Mouse'),
Has_child('Mammal', With_ancestor('Mouse', 'Mammal')),
Has_child(With_descendant('Mammal', 'Mouse'), 'Mouse'),
Whether_same('Mammal', 'Mammal', True),
Whether_same('Mammal', 'Meemmal', False),
Whether_same('Mouse', 'Mus', True),
Whether_same('Mus', 'Mouse', True),
Whether_same('Mus', 'Horse', False),
]
surprises = [
Has_child('Mouse', 'Mammal'),
Has_child('Mammal', 'Dog'), # fails
Whether_same('Mammal', 'Meemmal', True),
Whether_same('Mammal', 'Mammal', False),
]
make_claims(tax, new_claims)
passed = test_claims(tax, expectations)
passed = test_claims(tax, new_claims)
find_surprises(tax, surprises)
return passed
def test():
tax = Taxonomy.newTaxonomy()
tax.newTaxon('Mouse', 'species', 'about:blank')
tax.newTaxon('Dog', 'species', 'about:blank')
tax.newTaxon('Mammal', 'class', 'about:blank')
new_claims = [
Has_child('Mammal', 'Mouse', 'about:blank'),
Whether_same('Mouse', 'Mus', True),
Whether_same('Muus', 'Mouse', True),
]
expectations = [
Has_child('Mammal', 'Mouse'),
Has_child('Mammal', With_ancestor('Mouse', 'Mammal')),
Has_child(With_descendant('Mammal', 'Mouse'), 'Mouse'),
Whether_same('Mammal', 'Mammal', True),
Whether_same('Mammal', 'Meemmal', False),
Whether_same('Mouse', 'Mus', True),
Whether_same('Mus', 'Mouse', True),
Whether_same('Mus', 'Horse', False),
]
surprises = [
Has_child('Mouse', 'Mammal'),
Has_child('Mammal', 'Dog'), # fails
Whether_same('Mammal', 'Meemmal', True),
Whether_same('Mammal', 'Mammal', False),
]
make_claims(tax, new_claims)
passed = test_claims(tax, expectations)
passed = test_claims(tax, new_claims)
find_surprises(tax, surprises)
return passed
# Jython script to build the "model village" taxonomy.
from org.opentreeoflife.smasher import Taxonomy
# Create model taxonomy
tax = Taxonomy.newTaxonomy()
# 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/')
# analyzeOTUs sets flags on questionable taxa ("unclassified" and so on)
# to allow the option of suppression downstream
ncbi.analyzeOTUs()
tax.absorb(ncbi)
# Add GBIF subset fo the model taxonomy
gbif = Taxonomy.getTaxonomy('t/tax/gbif_aster/')
# 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()
tax.absorb(gbif)
# "Old" patch system with tab-delimited files
tax.edit('t/edits/')
# Example of referring to a taxon
fam = tax.taxon("Phellinaceae")
from org.opentreeoflife.smasher import Taxonomy
from org.opentreeoflife.smasher import Reportx
import taxonomies
ott = Taxonomy.newTaxonomy()
skel = Taxonomy.getTaxonomy('tax/skel/', 'skel')
ott.setSkeleton(skel)
def report(tax, tag):
ott.markDivisions(tax)
# Reportx.bogotypes(tax)
taxonomies.checkDivisions(tax)
Reportx.report(tax, tag + '-mrca-report.tsv')
if True:
report(taxonomies.loadIrmng(), 'irmng')
else:
silva = taxonomies.loadSilva()
ott.notSame(silva.taxon('Ctenophora', 'Coscinodiscophytina'),
skel.taxon('Ctenophora'))
report(silva, 'silva')
report(taxonomies.loadH2007(), 'h2007')
report(taxonomies.loadFung(), 'if')
report(taxonomies.loadNcbi(), 'ncbi')
report(taxonomies.loadGbif(), 'gbif')
report(taxonomies.loadIrmng(), 'irmng')
report(taxonomies.loadOtt(), 'ott')
# Jython script to build the Open Tree reference taxonomy
# coding=utf-8
# Unless specified otherwise issues are in the reference-taxonomy repo:
# https://github.com/OpenTreeOfLife/reference-taxonomy/issues/...
import sys
from org.opentreeoflife.smasher import Taxonomy
import taxonomies
sys.path.append("feed/misc/")
from chromista_spreadsheet import fixChromista
ott = Taxonomy.newTaxonomy()
skel = Taxonomy.getTaxonomy('tax/skel/', 'skel')
ott.setSkeleton(skel)
# ----- SILVA microbial taxonomy -----
def doSilva():
silva = taxonomies.loadSilva()
# - Deal with parent/child homonyms in SILVA -
# Arbitrary choices here to eliminate ambiguities down the road when NCBI gets merged.
# (If the homonym is retained, then the merge algorithm will have no
# way to choose between them, and refuse to match either. It will
# then create a third homonym.)
# Note order dependence between the following two
silva.taxon('Intramacronucleata','Intramacronucleata').rename('Intramacronucleata inf.')
silva.taxon('Spirotrichea','Intramacronucleata inf.').rename('Spirotrichea inf.')
silva.taxon('Cyanobacteria','Bacteria').rename('Cyanobacteria sup.')
# Jython script to build the "model village" taxonomy.
from org.opentreeoflife.smasher import Taxonomy
# Create model taxonomy
tax = Taxonomy.newTaxonomy()
# 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/')
# analyzeOTUs sets flags on questionable taxa ("unclassified" and so on)
# to allow the option of suppression downstream
ncbi.analyzeOTUs()
tax.absorb(ncbi)
# Add GBIF subset fo the model taxonomy
gbif = Taxonomy.getTaxonomy('t/tax/gbif_aster/')
# 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()
tax.absorb(gbif)
# "Old" patch system with tab-delimited files
tax.edit('t/edits/')
# Example of referring to a taxon
def create_ott():
ott = Taxonomy.newTaxonomy()
# There ought to be tests for all of these...
for name in names_of_interest:
ott.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 create_ott():
ott = Taxonomy.newTaxonomy()
# There ought to be tests for all of these...
for name in names_of_interest:
ott.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
