def test_add_domain_attribute():
test_data_dir = shephard.get_data('test_data')
fasta_file = '%s/%s' % (test_data_dir, 'testset_1.fasta')
domain_file = '%s/%s' % (test_data_dir, 'TS1_domains_idr.tsv')
P = uniprot.uniprot_fasta_to_proteome(fasta_file)
interfaces.si_domains.add_domains_from_file(P, domain_file)
prot = P.protein('O00401')
domain = prot.domains[0]
domain.add_attribute('test_attribute', 1)
assert domain.attribute('test_attribute') == 1
# this should fail
with pytest.raises(DomainException):
domain.add_attribute('test_attribute', 20)
# because the operation above should have failed, this too should
# have failed
assert domain.attribute('test_attribute') == 1
domain.add_attribute('test_attribute', 20, safe=False)
assert domain.attribute('test_attribute') == 20
assert len(domain.attributes) == 1
domain.add_attribute('another_test_attribute', 'testval')
assert len(domain.attributes) == 2
with pytest.raises(DomainException):
assert domain.attribute('does not exist') == 20
# check this returns none
assert domain.attribute('does not exist', safe=False) is None
def test_add_domains_file():
test_data_dir = shephard.get_data('test_data')
fasta_file = '%s/%s' % (test_data_dir, 'testset_1.fasta')
domain_file = '%s/%s' % (test_data_dir, 'TS1_domains_idr.tsv')
P = uniprot.uniprot_fasta_to_proteome(fasta_file)
interfaces.si_domains.add_domains_from_file(P, domain_file)
# this should fail because already added
with pytest.raises(ProteinException):
interfaces.si_domains.add_domains_from_file(P, domain_file)
P = uniprot.uniprot_fasta_to_proteome(fasta_file)
interfaces.si_domains.add_domains_from_file(P, domain_file, autoname=True)
print('')
P = uniprot.uniprot_fasta_to_proteome(fasta_file)
interfaces.si_domains.add_domains_from_file(P, domain_file, autoname=False)
# autoname allows 2 apparetly identical domain files to be added
interfaces.si_domains.add_domains_from_file(P, domain_file, autoname=True)
# autoname allows 2 apparetly identical domain files to be added
P = uniprot.uniprot_fasta_to_proteome(fasta_file)
interfaces.si_domains.add_domains_from_file(P,
domain_file,
autoname=False,
skip_bad=True)
def test_write_domain_with_attributes():
# this setup was also tested in test_add_domain_attribute
test_data_dir = shephard.get_data('test_data')
fasta_file = '%s/%s' % (test_data_dir, 'testset_1.fasta')
domain_file = '%s/%s' % (test_data_dir, 'TS1_domains_idr.tsv')
P = uniprot.uniprot_fasta_to_proteome(fasta_file)
interfaces.si_domains.add_domains_from_file(P, domain_file)
prot = P.protein('O00401')
domain = prot.domains[0]
domain.add_attribute('test_attribute_1', 1)
domain.add_attribute('test_attribute_cat', 'cat')
def test_fasta_to_proteome_part_1():
test_data_dir = shephard.get_data('test_data')
print(test_data_dir)
P = fasta.fasta_to_proteome('%s/%s' % (test_data_dir, 'testset_1.fasta'))
assert len(P.protein('1')) == 390
assert len(P) == 9
test_UID = 0
for i in P.proteins:
assert i == str(test_UID)
test_UID = test_UID + 1
P = fasta.fasta_to_proteome('%s/%s' % (test_data_dir, 'testset_1.fasta'),
proteome=P)
assert len(P) == 18
test_UID = 0
for i in P.proteins:
assert i == str(test_UID)
test_UID = test_UID + 1
##
## This block checks that removing a protein from the integer-indexed added proteins
## really removes it and that adding new proteins in does correctly start counting
## in the right place
P.remove_protein(10)
P = fasta.fasta_to_proteome('%s/%s' % (test_data_dir, 'testset_1.fasta'),
proteome=P)
test_UID = 0
for i in P.proteins:
if test_UID == 10:
with pytest.raises(ProteomeException):
assert P.protein(test_UID)
test_UID = test_UID + 1
assert i == str(test_UID)
test_UID = test_UID + 1
"""
import shephard
from shephard.interfaces import si_sites, si_domains, si_tracks, si_protein_attributes
from shephard.apis import uniprot
import pytest
import sys
TS1_FILE = [
'testset_1.fasta', 'TS1_domains_idr.tsv', 'TS1_domains_pscore.tsv',
'TS1_sites.tsv', 'TS1_tracks_pscore.tsv', 'TS1_protein_attributes.tsv',
'testset_1_ptms.tsv'
]
test_data_dir = shephard.get_data('test_data')
def build_proteome(fn):
return uniprot.uniprot_fasta_to_proteome('%s/%s' % (test_data_dir, fn))
@pytest.fixture
def TS1(request):
TS1_proteome = build_proteome(TS1_FILE[0])
return TS1_proteome
@pytest.fixture
def TS1_domains(request):
TS1_proteome = build_proteome(TS1_FILE[0])
def test_fasta_to_proteome_part_2():
def header_parser(s):
return s.split('|')[1]
test_data_dir = shephard.get_data('test_data')
print(test_data_dir)
# check we've read in the proteome OK
P = fasta.fasta_to_proteome('%s/%s' % (test_data_dir, 'testset_1.fasta'),
build_unique_ID=header_parser)
assert len(P) == 9
# this should trigger an exception because we're adding in duplicates
with pytest.raises(ProteomeException):
P = fasta.fasta_to_proteome('%s/%s' %
(test_data_dir, 'testset_1.fasta'),
build_unique_ID=header_parser,
proteome=P)
# this should not...
P = fasta.fasta_to_proteome('%s/%s' % (test_data_dir, 'testset_1.fasta'),
proteome=P)
assert len(P) == 18
# this also should not but should NOT add new sequences in
P = fasta.fasta_to_proteome('%s/%s' % (test_data_dir, 'testset_1.fasta'),
build_unique_ID=header_parser,
proteome=P,
force_overwrite=True)
assert len(P) == 18
P = fasta.fasta_to_proteome('%s/%s' % (test_data_dir, 'testset_1.fasta'),
proteome=P)
assert len(P) == 27
P = fasta.fasta_to_proteome('%s/%s' % (test_data_dir, 'testset_1.fasta'),
proteome=P,
use_header_as_unique_ID=True)
assert len(P) == 36
# this SHOULD trigger an exception because we shouldn't
with pytest.raises(APIException):
P = fasta.fasta_to_proteome('%s/%s' %
(test_data_dir, 'testset_1.fasta'),
proteome=P,
use_header_as_unique_ID=True,
build_unique_ID=header_parser)
expected_protein_uids = [
'O00401', 'O00470', 'O00472', 'O00499', 'O00629', 'O00712', 'O00716',
'O14786', 'Q9UJX3', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9',
'10', '11', '12', '13', '14', '15', '16', '17',
'sp|O00401|WASL_HUMAN Neural Wiskott-Aldrich syndrome protein OS=H**o sapiens OX=9606 GN=WASL PE=1 SV=2',
'sp|O00470|MEIS1_HUMAN Homeobox protein Meis1 OS=H**o sapiens OX=9606 GN=MEIS1 PE=1 SV=1',
'sp|O00472|ELL2_HUMAN RNA polymerase II elongation factor ELL2 OS=H**o sapiens OX=9606 GN=ELL2 PE=1 SV=2',
'sp|O00499|BIN1_HUMAN Myc box-dependent-interacting protein 1 OS=H**o sapiens OX=9606 GN=BIN1 PE=1 SV=1',
'sp|O00629|IMA3_HUMAN Importin subunit alpha-3 OS=H**o sapiens OX=9606 GN=KPNA4 PE=1 SV=1',
'sp|O00712|NFIB_HUMAN Nuclear factor 1 B-type OS=H**o sapiens OX=9606 GN=NFIB PE=1 SV=2',
'sp|O00716|E2F3_HUMAN Transcription factor E2F3 OS=H**o sapiens OX=9606 GN=E2F3 PE=1 SV=1',
'sp|O14786|NRP1_HUMAN Neuropilin-1 OS=H**o sapiens OX=9606 GN=NRP1 PE=1 SV=3',
'sp|Q9UJX3|APC7_HUMAN Anaphase-promoting complex subunit 7 OS=H**o sapiens OX=9606 GN=ANAPC7 PE=1 SV=4'
]
# CHECK all the unique IDs expected can be read in
for i in expected_protein_uids:
P.protein(i)
def test_add_protein():
# creating proteome and adding protein
test_data_dir = shephard.get_data('test_data')
P = uniprot.uniprot_fasta_to_proteome('%s/%s' %
(test_data_dir, 'testset_1.fasta'))
assert len(P.protein('O00401')) == 505
assert len(P.protein('O00470')) == 390
assert len(P.protein('O00472')) == 640
assert len(P.protein('O00499')) == 593
assert len(P.protein('O00629')) == 521
assert len(P.protein('O00712')) == 420
assert len(P.protein('O00716')) == 465
assert len(P.protein('O14786')) == 923
assert len(P.protein('Q9UJX3')) == 599
# creating a proteome from a FASTA file (using defaul unique key)
P = fasta.fasta_to_proteome('%s/%s' % (test_data_dir, 'testset_1.fasta'))
assert len(P.protein('1')) == 390
assert len(P.protein('2')) == 640
assert len(P.protein('3')) == 593
assert len(P.protein('4')) == 521
assert len(P.protein('5')) == 420
assert len(P.protein('6')) == 465
assert len(P.protein('7')) == 923
assert len(P.protein('8')) == 599
# create a proteome where FASTA header is used as uniqueID
P = fasta.fasta_to_proteome('%s/%s' % (test_data_dir, 'testset_1.fasta'),
use_header_as_unique_ID=True)
assert len(
P.protein(
'sp|O00470|MEIS1_HUMAN Homeobox protein Meis1 OS=H**o sapiens OX=9606 GN=MEIS1 PE=1 SV=1'
)) == 390
assert len(
P.protein(
'sp|O00472|ELL2_HUMAN RNA polymerase II elongation factor ELL2 OS=H**o sapiens OX=9606 GN=ELL2 PE=1 SV=2'
)) == 640
assert len(
P.protein(
'sp|O00499|BIN1_HUMAN Myc box-dependent-interacting protein 1 OS=H**o sapiens OX=9606 GN=BIN1 PE=1 SV=1'
)) == 593
assert len(
P.protein(
'sp|O00629|IMA3_HUMAN Importin subunit alpha-3 OS=H**o sapiens OX=9606 GN=KPNA4 PE=1 SV=1'
)) == 521
assert len(
P.protein(
'sp|O00712|NFIB_HUMAN Nuclear factor 1 B-type OS=H**o sapiens OX=9606 GN=NFIB PE=1 SV=2'
)) == 420
assert len(
P.protein(
'sp|O00716|E2F3_HUMAN Transcription factor E2F3 OS=H**o sapiens OX=9606 GN=E2F3 PE=1 SV=1'
)) == 465
assert len(
P.protein(
'sp|O14786|NRP1_HUMAN Neuropilin-1 OS=H**o sapiens OX=9606 GN=NRP1 PE=1 SV=3'
)) == 923
assert len(
P.protein(
'sp|Q9UJX3|APC7_HUMAN Anaphase-promoting complex subunit 7 OS=H**o sapiens OX=9606 GN=ANAPC7 PE=1 SV=4'
)) == 599
# check manually adding proteomes
local_seq = 'PPPPP'
P.add_protein(local_seq, '5pp', 'U5P')
assert P.protein('U5P').sequence == local_seq
assert P.protein('U5P').name == '5pp'
# should trigger exception
with pytest.raises(ProteomeException):
P.add_protein(local_seq, '5pp', 'U5P')
P.add_protein('ASDF', '5pp', 'U5P', force_overwrite=True)
assert P.protein('U5P').sequence == 'ASDF'
protein_list = []
p1 = {
'sequence': 'ASDFGH',
'name': "Test protein 1",
'unique_ID': 1.23,
"attributes": None
}
protein_list.append(p1)
# check this works
P = proteome.Proteome(protein_list)
print(P.proteins)
assert P.protein(1.23).sequence == 'ASDFGH'
assert P.protein("1.23").sequence == 'ASDFGH'
P.remove_protein(1.23)
with pytest.raises(ProteomeException):
assert P.protein(1.23).sequence == 'ASDFGH'