def abundance_dict(species):
"""Returns dict - uniprot => (decay, abundance)"""
header, data = utils.load_ned_data(species)
pi = header.index('proID')
ai = header.index('rel.abun')
di = header.index('def')
return {line[pi]: (float(line[ai]), line[di]) for line in data}
def NED_core_interactor_test(protein_subset=None):
header, data = utils.load_ned_data('human')
# data = [line for line in data if line[0] in protein_subset]
neds = set(line[-2] for line in data if line[-3] == 'NED')
eds = set(line[-2] for line in data if line[-3] == 'ED')
header, data = load_hein()
baits = set(protein for line in data for protein in line[2])
# Select set of proteins common to both NED data and stoich data.
# baits = baits.intersection(neds.union(eds))
# neds = neds.intersection(baits)
# eds = eds.intersection(baits)
# Test likelihood of finding NED proteins in Core
print(len(neds), len(eds))
ned_core, ed_core, ned_noncore, ed_noncore = 0, 0, 0, 0
for line in data:
# Ignore self-interactors
if line[14] == '+' or line[15] == '+':
continue
if line[2].intersection(neds) != set():
if line[13] == '+':
ned_core += 1
else:
ned_noncore += 1
elif line[2].intersection(eds) != set():
if line[13] == '+':
ed_core += 1
else:
ed_noncore += 1
print(ned_core, ed_core, ned_noncore, ed_noncore)
def load_paxdb_data(species):
if species == 'mouse':
meta = paxdb.get_metadata('10090')
elif species == 'human':
meta = paxdb.get_metadata('9606')
data = utils.load_ned_data(species)[1]
pmap = protein_map(species)
skip = ['CELL_LINE']
for i in skip:
meta.pop(i)
abunds = [paxdb.Abundances(meta[t]['filename'], 0) for t in sorted(meta)]
header = '\t'.join(['prot', '\t'.join(sorted(meta)), 'tcount', 'def'])
return header, data, abunds, pmap
def __init__(self, species, homologs=False):
"""Mouse specific if using homologs, else human or mouse."""
neds = utils.load_ned_data(species)[1]
if homologs == True:
if species != 'mouse':
raise ValueError('homologs only implemented for mouse')
self._homologs = utils.get_homologs()
self._corum = dbloader.LoadCorum(version='core')
else:
self._homologs = False
self._corum = dbloader.LoadCorum(species, 'core')
self._coex = coexpressdb.Coexpression(species)
# Check these indices if using new datasets
self._decay = {line[0]: line[-2] for line in neds}
self._species = species
self._outdata = []
def load_proteomicsdb_data():
data = utils.load_ned_data('human')[1]
isoab = proteomicsdb.Abundances('human_protdb_isoforms_expression.txt')
abunds = proteomicsdb.Abundances('trembl_tissues.txt')
tissues = abunds.tissues
header = '{0}\t{1}\t{2}\t{3}\n'.format('prot', '\t'.join(tissues),
'tcount', 'def')
with open('data/human_proteomicsdb_tissue_expression.txt', 'w') as outfile:
outfile.write(header)
for line in data:
prot = line[0]
if prot in abunds.proteins:
expression = [abunds.expression(prot, tis) for tis in ttargets]
elif prot in isoab.proteins:
expression = [isoab.expression(prot, tis) for tis in ttargets]
else:
continue
tcount = str(len(expression) - expression.count('NA'))
newline = '\t'.join([prot] + expression + [tcount, line[-1]])
outfile.write(newline + '\n')
def test_abundance_dict(species):
if species == 'mouse_homologs':
return
_, data = utils.load_ned_data(species)
abdict = ab.abundance_dict(species)
assert len(data) == len(abdict)
def test_load_ned_data(species):
header, data = utils.load_ned_data(species)
for colname in ['proID', 'Rel. abund', 'def']:
assert colname in header
assert len(data[0]) == len(header)