def liftover_mods(ref_seq, ref_nc_seq, seq, can_monomers=can_monomers):
ref_form = bpforms.RnaForm().from_str(ref_nc_seq)
form = bpforms.RnaForm()
i_nc_nt = 0
for i_nt, ref_monomer in enumerate(ref_seq):
if i_nc_nt < len(ref_form.seq):
ref_nc_monomer = ref_form.seq[i_nc_nt]
else:
ref_nc_monomer = None
if ref_monomer != '-':
i_nc_nt += 1
monomer = seq[i_nt]
if monomer == '-':
continue
elif ref_nc_monomer not in can_monomers and monomer == ref_monomer:
form.seq.append(ref_nc_monomer)
elif monomer == 'N':
form.seq.append(bpforms.Monomer(id='N'))
else:
form.seq.append(bpforms.rna_alphabet.monomers.get(monomer))
# verify non-canonical sequences are consistent with the canonical sequences
assert seq.replace('-', '') == form.get_canonical_seq()
return form
def analyze_form(rna_form, unsupported_codes, results_dict):
results_dict['Sequence (BpForms)'] = str(rna_form)
results_dict['Sequence (IUPAC)'] = canonical_seq = rna_form.get_canonical_seq()
results_dict['Length'] = len(rna_form.seq)
results_dict['Number of modifications'] = len(rna_form.seq) \
- rna_form.seq.count(bpforms.rna_alphabet.monomers.A) \
- rna_form.seq.count(bpforms.rna_alphabet.monomers.C) \
- rna_form.seq.count(bpforms.rna_alphabet.monomers.G) \
- rna_form.seq.count(bpforms.rna_alphabet.monomers.U)
results_dict['Number of modified A'] = canonical_seq.count('A') - rna_form.seq.count(bpforms.rna_alphabet.monomers.A)
results_dict['Number of modified C'] = canonical_seq.count('C') - rna_form.seq.count(bpforms.rna_alphabet.monomers.C)
results_dict['Number of modified G'] = canonical_seq.count('G') - rna_form.seq.count(bpforms.rna_alphabet.monomers.G)
results_dict['Number of modified U'] = canonical_seq.count('U') - rna_form.seq.count(bpforms.rna_alphabet.monomers.U)
if unsupported_codes:
results_dict['BpForms errors'] = 'MODOMICS sequence uses monomeric forms {}'.format(
', '.join(unsupported_codes))
else:
results_dict['Formula'] = str(rna_form.get_formula())
results_dict['Molecular weight'] = rna_form.get_mol_wt()
results_dict['Charge'] = rna_form.get_charge()
canonical_form = bpforms.RnaForm().from_str(canonical_seq)
results_dict['Canonical formula'] = str(canonical_form.get_formula())
results_dict['Canonical molecular weight'] = canonical_form.get_mol_wt()
results_dict['Canonical charge'] = canonical_form.get_charge()
results_dict['Extra formula'] = str(rna_form.get_formula() - canonical_form.get_formula())
results_dict['Extra molecular weight'] = rna_form.get_mol_wt() - canonical_form.get_mol_wt()
results_dict['Extra charge'] = rna_form.get_charge() - canonical_form.get_charge()
results_dict['BpForms errors'] = ' '.join(rna_form.validate())
def test_serialize_bpforms(self):
seq = 'ACGU'
mol = bpforms.RnaForm().from_str(seq)
s = obj_tables.chem.ChemicalStructure(mol)
self.assertEqual(s.serialize(),
'{}/{}: {}'.format('bpforms', 'rna', seq))
tot_rna = 375000
doubling_time = 45. # min
half_life = 45. # min
monomer_freq = {}
can_monomer_freq = {}
monomer_codes = {
monomer: code
for code, monomer in bpforms.rna_alphabet.monomers.items()
}
tot_copies = 0
with open('examples/modomics_trna_copy_numbers.csv', 'r') as file:
for rna in csv.DictReader(file, dialect='excel'):
form = bpforms.RnaForm().from_str(rna['Sequence (BpForms)'])
copies = float(rna['Copies per cell'])
tot_copies += copies
for monomer in form.seq:
if monomer not in canonical_monomers:
if monomer not in monomer_freq:
monomer_freq[monomer] = 0
monomer_freq[monomer] += copies
can_code = monomer.get_canonical_code(monomer_codes)
if can_code not in can_monomer_freq:
can_monomer_freq[can_code] = 0
can_monomer_freq[can_code] += copies
# verify non-canonical sequences are consistent with the canonical sequences
assert seq.replace('-', '') == form.get_canonical_seq()
return form
'''
rRNA
'''
rrna_types = ['5.8S', '18S', '28S']
for rrna_type in rrna_types:
filename = 'examples/homo_sapiens_rna/{} all seqs.fasta'.format(rrna_type)
seqs = [str(record.seq) for record in SeqIO.parse(filename, "fasta")]
ref_nc_seq = seqs[0]
ref_form = bpforms.RnaForm().from_str(ref_nc_seq)
ref_seq = seqs[1]
seqs = seqs[1:]
# map curated modifications onto sequences
forms = []
for seq in seqs:
forms.append(liftover_mods(ref_seq, ref_nc_seq, seq))
# save non-canonical sequences
with open(
'examples/homo_sapiens_rna/{} nc alignment.txt'.format(rrna_type),
'w') as file:
for form in forms:
file.write(str(form) + '\n')
'''
def run_rrna(session, modomics_short_code_to_monomer, monomer_codes, out_filename):
response = session.get(URL, params={
'RNA_type': 'rRNA',
'RNA_subtype': 'all',
'organism': 'all species',
'vis_type': 'Modomics symbols',
})
response.raise_for_status()
doc = bs4.BeautifulSoup(response.text, 'lxml')
table = doc.find('table', {'id': 'tseq'})
tbody = table.find('tbody')
rows = tbody.find_all('tr')
rna_forms = []
for row in rows:
if not isinstance(row, bs4.element.Tag):
continue
cells = row.find_all('td')
rna_form = bpforms.RnaForm()
unsupported_codes = set()
for child in cells[5].children:
if child.name is None or child.name == 'span':
if child.name is None:
text = str(child)
else:
text = child.text
for code in text.strip().replace('-', '').replace('_', ''):
monomer = modomics_short_code_to_monomer.get(code, None)
if monomer is None:
unsupported_codes.add(code)
monomer = bpforms.Monomer(id=code)
else:
monomer_codes[code] = monomer
rna_form.seq.append(monomer)
elif child.name == 'a':
code = child.get('href').replace('/modomics/modifications/', '')
monomer = modomics_short_code_to_monomer.get(code, None)
if monomer is None:
unsupported_codes.add(code)
monomer = bpforms.Monomer(id=code)
else:
monomer_codes[code] = monomer
rna_form.seq.append(monomer)
else:
raise Exception('Unsupported child {}'.format(child.name))
rna_forms.append({
'GenBank': cells[0].find('a').text.strip(),
'Organism': cells[3].text.strip(),
'Organellum': cells[4].text.strip(),
'Type': cells[2].text.strip(),
'Sequence (MODOMICS)': cells[5].text.strip().replace('-', '').replace('_', ''),
})
analyze_form(rna_form, unsupported_codes, rna_forms[-1])
# save results to tab-separated file
save_results(rna_forms, ['GenBank', 'Type'], out_filename)
return rna_forms
def run_trna(session, modomics_short_code_to_monomer, monomer_codes, out_filename):
response = session.get(URL, params={
'RNA_type': 'tRNA',
'RNA_subtype': 'all',
'organism': 'all species',
'vis_type': 'Modomics symbols',
})
response.raise_for_status()
doc = bs4.BeautifulSoup(response.text, 'lxml')
table = doc.find('table', {'id': 'tseq'})
tbody = table.find('tbody')
rows = tbody.find_all('tr')
rna_forms = []
code_freq = {}
canonical_code_freq = {'A': 0, 'C': 0, 'G': 0, 'U': 0}
for row in rows:
cells = row.find_all('td')
rna_form = bpforms.RnaForm()
unsupported_codes = set()
for child in cells[5].children:
if child.name is None or child.name == 'span':
if child.name is None:
text = str(child)
else:
text = child.text
for code in text.strip().replace('-', '').replace('_', ''):
monomer = modomics_short_code_to_monomer.get(code, None)
if monomer is None:
unsupported_codes.add(code)
monomer = bpforms.Monomer(id=code)
else:
monomer_codes[code] = monomer
if code not in code_freq:
code_freq[code] = 0
code_freq[code] += 1
rna_form.seq.append(monomer)
elif child.name == 'a':
code = child.get('href').replace('/modomics/modifications/', '')
monomer = modomics_short_code_to_monomer.get(code, None)
if monomer is None:
unsupported_codes.add(code)
monomer = bpforms.Monomer(id=code)
else:
monomer_codes[code] = monomer
if code not in code_freq:
code_freq[code] = 0
code_freq[code] += 1
rna_form.seq.append(monomer)
else:
raise Exception('Unsupported child {}'.format(child.name))
rna_forms.append({
'Amino acid type': cells[1].text.strip(),
'Anticodon': cells[2].text.strip(),
'Organism': cells[3].text.strip(),
'Organellum': cells[4].text.strip(),
'Sequence (MODOMICS)': cells[5].text.strip().replace('-', '').replace('_', ''),
})
analyze_form(rna_form, unsupported_codes, rna_forms[-1])
canonical_code_freq['A'] += \
rna_forms[-1]['Sequence (IUPAC)'].count('A') \
- rna_form.seq.count(bpforms.rna_alphabet.monomers.A)
canonical_code_freq['C'] += \
rna_forms[-1]['Sequence (IUPAC)'].count('C') \
- rna_form.seq.count(bpforms.rna_alphabet.monomers.C)
canonical_code_freq['G'] += \
rna_forms[-1]['Sequence (IUPAC)'].count('G') \
- rna_form.seq.count(bpforms.rna_alphabet.monomers.G)
canonical_code_freq['U'] += \
rna_forms[-1]['Sequence (IUPAC)'].count('U') \
- rna_form.seq.count(bpforms.rna_alphabet.monomers.U)
# save results to tab-separated file
save_results(rna_forms, ['Amino acid type', 'Anticodon'], out_filename)
with open(os.path.join('examples', 'modomics.trna.canonical-code-freq.tsv'), 'w') as file:
writer = csv.DictWriter(file, fieldnames=['Code', 'Frequency'], dialect='excel-tab')
writer.writeheader()
for code, freq in canonical_code_freq.items():
writer.writerow({'Code': code, 'Frequency': freq})
with open(os.path.join('examples', 'modomics.trna.code-freq.tsv'), 'w') as file:
writer = csv.DictWriter(file, fieldnames=['Code', 'Frequency'], dialect='excel-tab')
writer.writeheader()
for code, freq in code_freq.items():
writer.writerow({'Code': code, 'Frequency': freq})
return rna_forms, canonical_code_freq, code_freq
def post(self):
ret = {}
warnings = []
args = bcform_ns.payload
# print(args)
# get arguments
form = args['form']
arg_subunits = args.get('subunits', None)
# validate form
try:
bc_form = bcforms.core.BcForm().from_str(form)
except Exception as error:
flask_restplus.abort(400,
'Form is invalid',
errors={'form': str(error)})
errors = bc_form.validate()
if errors:
flask_restplus.abort(400,
'Form is invalid',
errors={'form': '. '.join(errors)})
# validate input subunit properties
sum_length = 0
if arg_subunits is not None:
for subunit in arg_subunits:
# check if name is in the form
subunit_id = subunit['name']
if subunit_id in [subunit.id for subunit in bc_form.subunits]:
# check if encoding and structure are present at the same time
if ('encoding' in subunit) and ('structure' in subunit):
# if encoding and structure both present, check if encoding is known
encoding = subunit['encoding'].strip()
if encoding == 'bpforms.ProteinForm':
try:
subunit_structure = bpforms.ProteinForm(
).from_str(subunit['structure'])
sum_length += len(
subunit_structure
) * bc_form.get_subunit_attribute(
subunit_id, 'stoichiometry')
bc_form.set_subunit_attribute(
subunit_id, 'structure', subunit_structure)
except Exception as error:
flask_restplus.abort(
400,
'Unable to parse bpforms.ProteinForm',
errors={'structure': str(error)})
elif encoding == 'bpforms.DnaForm':
try:
subunit_structure = bpforms.DnaForm().from_str(
subunit['structure'])
sum_length += len(
subunit_structure
) * bc_form.get_subunit_attribute(
subunit_id, 'stoichiometry')
bc_form.set_subunit_attribute(
subunit_id, 'structure', subunit_structure)
except Exception as error:
flask_restplus.abort(
400,
'Unable to parse bpforms.DnaForm',
errors={'structure': str(error)})
elif encoding == 'bpforms.RnaForm':
try:
subunit_structure = bpforms.RnaForm().from_str(
subunit['structure'])
sum_length += len(
subunit_structure
) * bc_form.get_subunit_attribute(
subunit_id, 'stoichiometry')
bc_form.set_subunit_attribute(
subunit_id, 'structure', subunit_structure)
except Exception as error:
flask_restplus.abort(
400,
'Unable to parse bpforms.RnaForm',
errors={'structure': str(error)})
elif encoding == 'smiles' or encoding == 'SMILES' or encoding == 'smi' or encoding == 'SMI':
try:
bc_form.set_subunit_attribute(
subunit_id, 'structure',
subunit['structure'])
except Exception as error:
flask_restplus.abort(
400,
'Unable to parse SMILES string',
errors={'structure': str(error)})
# else if one is present but not the other, report error
elif ('encoding' in subunit) ^ ('structure' in subunit):
flask_restplus.abort(
400,
'One of encoding and structure is present but not both'
)
# when neither encoding nor structure is present
else:
# check formula
if 'formula' in subunit:
try:
bc_form.set_subunit_attribute(
subunit_id, 'formula', subunit['formula'])
except Exception as error:
flask_restplus.abort(
400,
'Unable to parse formula',
errors={'formula': str(error)})
elif 'mol_wt' in subunit:
try:
bc_form.set_subunit_attribute(
subunit_id, 'mol_wt', subunit['mol_wt'])
except Exception as error:
flask_restplus.abort(
400,
'Unable to parse mol_wt',
errors={'mol_wt': str(error)})
# check charge
if 'charge' in subunit:
try:
bc_form.set_subunit_attribute(
subunit_id, 'charge', subunit['charge'])
except Exception as error:
flask_restplus.abort(
400,
'Unable to parse charge',
errors={'charge': str(error)})
else:
flask_restplus.abort(400,
'Subunit name not in BcForm',
errors={'subunit': subunit_id})
ret['form'] = str(bc_form)
if sum_length <= max_len_get_structure:
try:
ret['structure'] = bc_form.export()
except Exception:
pass
else:
warnings.append(
'The sum of length of bpforms-encoded subunits is {}, which exceeds the max length limit {}.'
.format(sum_length, max_len_get_structure))
ret['structure'] = None
try:
ret['formula'] = str(bc_form.get_formula())
except Exception:
pass
try:
ret['mol_wt'] = bc_form.get_mol_wt()
except Exception:
pass
try:
ret['charge'] = bc_form.get_charge()
except Exception:
pass
if len(warnings) > 0:
ret['warnings'] = ' '.join(warnings)
return ret