def test_remove_genes() -> None:
"""Test gene removal."""
m = Model("test")
m.add_reactions([Reaction("r" + str(i + 1)) for i in range(8)])
assert len(m.reactions) == 8
rxns = m.reactions
rxns.r1.gene_reaction_rule = "(a and b) or (c and a)"
rxns.r2.gene_reaction_rule = "(a and b and d and e)"
rxns.r3.gene_reaction_rule = "(a and b) or (b and c)"
rxns.r4.gene_reaction_rule = "(f and b) or (b and c)"
rxns.r5.gene_reaction_rule = "x"
rxns.r6.gene_reaction_rule = "y"
rxns.r7.gene_reaction_rule = "x or z"
rxns.r8.gene_reaction_rule = ""
assert "a" in m.genes
assert "x" in m.genes
remove_genes(m, ["a"], remove_reactions=False)
assert "a" not in m.genes
assert "x" in m.genes
assert rxns.r1.gene_reaction_rule == ""
assert rxns.r2.gene_reaction_rule == ""
assert rxns.r3.gene_reaction_rule == "b and c"
assert rxns.r4.gene_reaction_rule == "(f and b) or (b and c)"
assert rxns.r5.gene_reaction_rule == "x"
assert rxns.r6.gene_reaction_rule == "y"
assert rxns.r7.genes == {m.genes.x, m.genes.z}
assert rxns.r8.gene_reaction_rule == ""
remove_genes(m, ["x"], remove_reactions=True)
assert len(m.reactions) == 7
assert "r5" not in m.reactions
assert "x" not in m.genes
assert rxns.r1.gene_reaction_rule == ""
assert rxns.r2.gene_reaction_rule == ""
assert rxns.r3.gene_reaction_rule == "b and c"
assert rxns.r4.gene_reaction_rule == "(f and b) or (b and c)"
assert rxns.r6.gene_reaction_rule == "y"
assert rxns.r7.gene_reaction_rule == "z"
assert rxns.r7.genes == {m.genes.z}
assert rxns.r8.gene_reaction_rule == ""
def getModelWithRemoveGene(model0, gene_remove0):
model1 = model0.copy()
remove_genes(model1, gene_remove0,
remove_reactions=True) # this script can't be used
return model1
def convert_ids(model):
"""Converts metabolite and reaction ids to the new style.
Returns a tuple with the new model and a dictionary of old ids set up like this:
{'reactions': {'new_id': 'old_id'},
'metabolites': {'new_id': 'old_id'},
'genes': {'new_id': 'old_id'}}
"""
# loop through the ids:
metabolite_id_dict = defaultdict(list)
reaction_id_dict = defaultdict(list)
gene_id_dict = defaultdict(list)
# fix metabolites
for metabolite in model.metabolites:
new_id = id_for_new_id_style(fix_legacy_id(metabolite.id,
use_hyphens=False),
is_metabolite=True)
metabolite_id_dict[new_id].append(metabolite.id)
metabolite.id = new_id
model.metabolites._generate_index()
# take out the _b metabolites
remove_boundary_metabolites(model)
# load fixes for gene_reaction_rule's
rule_prefs = _get_rule_prefs()
# separate ids and compartments, and convert to the new_id_style
for reaction in model.reactions:
# save the original id
current_id = reaction.id
# apply new id style
reaction.id = id_for_new_id_style(
fix_legacy_id(reaction.id, use_hyphens=False))
# normalize pseudoreaction IDs
try:
_normalize_pseudoreaction(reaction)
except ConflictingPseudoreaction as e:
logging.warn(str(e))
# don't merge reactions with conflicting new_id's
while reaction.id in reaction_id_dict:
reaction.id = increment_id(reaction.id)
reaction_id_dict[reaction.id].append(current_id)
# fix the gene reaction rules
reaction.gene_reaction_rule = _check_rule_prefs(
rule_prefs, reaction.gene_reaction_rule)
model.reactions._generate_index()
# update the genes
for gene in list(model.genes):
new_id = scrub_gene_id(gene.id)
gene_id_dict[new_id].append(gene.id)
for reaction in gene.reactions:
reaction.gene_reaction_rule = re.sub(
r'\b' + re.escape(gene.id) + r'\b', new_id,
reaction.gene_reaction_rule)
# remove old genes
from cobra.manipulation import remove_genes
remove_genes(model,
[gene for gene in model.genes if len(gene.reactions) == 0])
# fix the model id
cobra_id = re.sub(r'[^a-zA-Z0-9_]', '_', model.id)
model.id = cobra_id
old_ids = {
'metabolites': metabolite_id_dict,
'reactions': reaction_id_dict,
'genes': gene_id_dict
}
return model, old_ids
def convert_ids(model):
"""Converts metabolite and reaction ids to the new style.
Returns a tuple with the new model and a dictionary of old ids set up like this:
{'reactions': {'new_id': 'old_id'},
'metabolites': {'new_id': 'old_id'},
'genes': {'new_id': 'old_id'}}
"""
# loop through the ids:
metabolite_id_dict = defaultdict(list)
reaction_id_dict = defaultdict(list)
gene_id_dict = defaultdict(list)
# fix metabolites
for metabolite in model.metabolites:
new_id = id_for_new_id_style(fix_legacy_id(metabolite.id, use_hyphens=False),
is_metabolite=True)
metabolite_id_dict[new_id].append(metabolite.id)
if new_id != metabolite.id:
# new_id already exists, then merge
if new_id in model.metabolites:
new_id = add_duplicate_tag(new_id)
while new_id in model.metabolites:
new_id = increment_id(new_id)
metabolite.id = new_id
model.metabolites._generate_index()
# take out the _b metabolites
remove_boundary_metabolites(model)
# load fixes for gene_reaction_rule's
rule_prefs = _get_rule_prefs()
# separate ids and compartments, and convert to the new_id_style
for reaction in model.reactions:
# apply new id style
new_style_id = id_for_new_id_style(fix_legacy_id(reaction.id, use_hyphens=False))
# normalize pseudoreaction IDs
try:
pseudo_id = _normalize_pseudoreaction(new_style_id, reaction)
except ConflictingPseudoreaction as e:
logging.warn(str(e))
new_id = pseudo_id if pseudo_id is not None else new_style_id
# don't merge reactions with conflicting new_id's
if new_id != reaction.id and new_id in model.reactions:
new_id = add_duplicate_tag(new_id)
while new_id in model.reactions:
new_id = increment_id(new_id)
reaction_id_dict[new_id].append(reaction.id)
reaction.id = new_id
# fix the gene reaction rules
reaction.gene_reaction_rule = _check_rule_prefs(rule_prefs, reaction.gene_reaction_rule)
model.reactions._generate_index()
# update the genes
for gene in list(model.genes):
new_id = scrub_gene_id(gene.id)
gene_id_dict[new_id].append(gene.id)
for reaction in gene.reactions:
reaction.gene_reaction_rule = re.sub(r'\b' + re.escape(gene.id) + r'\b', new_id,
reaction.gene_reaction_rule)
# remove old genes
from cobra.manipulation import remove_genes
remove_genes(model, [gene for gene in model.genes
if len(gene.reactions) == 0])
# fix the model id
bigg_id = re.sub(r'[^a-zA-Z0-9_]', '_', model.id)
model.id = bigg_id
old_ids = {'metabolites': metabolite_id_dict,
'reactions': reaction_id_dict,
'genes': gene_id_dict}
return model, old_ids