def parse_group_sbml(groups_sbml, chebi):
doc = libsbml.SBMLReader().readSBMLFromFile(groups_sbml)
groups_model = doc.getModel()
groups_plugin = groups_model.getPlugin("groups")
r_id2g_id, s_id2gr_id, ub_sps = {}, {}, set()
if groups_plugin:
for group in groups_plugin.getListOfGroups():
gr_members = [it.getIdRef() for it in group.getListOfMembers()]
gr_id, gr_name = group.getId(), group.getName()
gr_sbo = group.getSBOTermID()
gr_type = next(get_annotations(group, libsbml.BQB_IS_DESCRIBED_BY), None)
if not gr_type:
continue
if SBO_BIOCHEMICAL_REACTION == gr_sbo:
if GROUP_TYPE_EQUIV == gr_type:
for r_id in gr_members:
r_id2g_id[r_id] = gr_id, gr_name, len(gr_members)
elif SBO_CHEMICAL_MACROMOLECULE == gr_sbo:
if GROUP_TYPE_UBIQUITOUS == gr_type:
ub_sps = set(gr_members)
elif GROUP_TYPE_EQUIV == gr_type:
for s_id in gr_members:
term = get_chebi_term_by_annotation(group, chebi)
s_id2gr_id[s_id] = gr_id, term if term else gr_name, len(gr_members)
else:
raise GrPlError()
return r_id2g_id, s_id2gr_id, ub_sps
def parse_group_sbml(groups_sbml, chebi):
doc = libsbml.SBMLReader().readSBMLFromFile(groups_sbml)
groups_model = doc.getModel()
groups_plugin = groups_model.getPlugin("groups")
r_id2g_id, s_id2gr_id, ub_sps = {}, {}, set()
if groups_plugin:
for group in groups_plugin.getListOfGroups():
gr_members = [it.getIdRef() for it in group.getListOfMembers()]
gr_id, gr_name = group.getId(), group.getName()
gr_sbo = group.getSBOTermID()
gr_type = next(get_annotations(group, libsbml.BQB_IS_DESCRIBED_BY), None)
if not gr_type:
continue
if SBO_BIOCHEMICAL_REACTION == gr_sbo:
if GROUP_TYPE_EQUIV == gr_type:
for r_id in gr_members:
r_id2g_id[r_id] = gr_id, gr_name, len(gr_members)
elif SBO_CHEMICAL_MACROMOLECULE == gr_sbo:
if GROUP_TYPE_UBIQUITOUS == gr_type:
ub_sps = set(gr_members)
elif GROUP_TYPE_EQUIV == gr_type:
for s_id in gr_members:
term = get_chebi_term_by_annotation(group, chebi)
s_id2gr_id[s_id] = gr_id, term if term else gr_name, len(gr_members)
else:
raise GrPlError()
return r_id2g_id, s_id2gr_id, ub_sps
def serialize_generalization(r_id2clu, s_id2clu, sbml, chebi, path):
doc = libsbml.SBMLReader().readSBML(sbml)
model = doc.getModel()
groups_plugin = model.getPlugin("groups")
clu2r_ids, clu2s_ids = invert_map(r_id2clu), invert_map(s_id2clu)
wb = openpyxl.Workbook()
ws = wb.create_sheet(0, "Metabolite Groups")
row = 1
add_values(ws, row, 1, ["Group Id", "Group Name", "Group CHEBI", "Id", "Name", "Compartment", "CHEBI"], HEADER_STYLE)
row += 1
processed_s_ids = set()
for (g_id, ch_term), s_ids in sorted(clu2s_ids.items(), key=lambda ((g_id, _), s_ids): g_id):
group = groups_plugin.getGroup(g_id)
add_values(ws, row, 1, [g_id, group.getName(), ch_term.get_id() if ch_term else ''])
for s_id in sorted(s_ids, key=lambda s_id: s_id[s_id.find('__'):]):
species = model.getSpecies(s_id)
ch_term = get_chebi_term_by_annotation(species, chebi)
add_values(ws, row, 4, [s_id, species.getName(), model.getCompartment(species.getCompartment()).getName(),
ch_term.get_id() if ch_term else ''])
row += 1
processed_s_ids |= s_ids
add_values(ws, row, 1, [g_id, group.getName(), ch_term.get_id() if ch_term else ''])
for s_id in sorted(s_ids, key=lambda s_id: s_id[s_id.find('__'):]):
species = model.getSpecies(s_id)
ch_term = get_chebi_term_by_annotation(species, chebi)
add_values(ws, row, 4, [s_id, species.getName(), model.getCompartment(species.getCompartment()).getName(),
ch_term.get_id() if ch_term else ''])
row += 1
processed_s_ids |= s_ids
ws = wb.create_sheet(1, "Ungrouped metabolites")
row = 1
add_values(ws, row, 1, ["Id", "Name", "Compartment", "CHEBI"], HEADER_STYLE)
row += 1
unm_l = 0
for species in sorted(model.getListOfSpecies(), key=lambda s: s.getId()[s.getId().find('__'):]):
if species.getId() not in processed_s_ids:
ch_term = get_chebi_term_by_annotation(species, chebi)
add_values(ws, row, 1, [species.getId(), species.getName(),
model.getCompartment(species.getCompartment()).getName(),
ch_term.get_id() if ch_term else ''])
row += 1
unm_l += 1
print unm_l
ws = wb.create_sheet(2, "Reaction Groups")
row = 1
add_values(ws, row, 1, ["Group Id", "Id", "Name", "Formula", "Gene Association"], HEADER_STYLE)
row += 1
processed_r_ids = set()
for (g_id, g_name), r_ids in sorted(clu2r_ids.items(), key=lambda ((g_id, g_name), _): g_id):
add_values(ws, row, 1, [g_id])
for r_id in sorted(r_ids, key=lambda r_id: r_id[r_id.find('__'):]):
