def seqvar_or_None(entry):
if entry is None:
return None
# basic string cleaning
entry = entry.strip()
# beware the pesky N-dash in at least 1 observed ClinVar db entry.
entry = entry.replace('\u2013', '-')
hgvs_text = ''
match = re_hgvs.match(entry)
if match:
stuff = match.groupdict()
hgvs_text = stuff['prefix'] + '_' + stuff['transcript'] + ':' + stuff[
'edit']
# TODO: LRG
else:
return None
try:
return Variant(hgvs_text)
except (CriticalHgvsError, TypeError):
# empty or broken
return None
def test_simple_substitution_case(self):
var_c = Variant(hgvs_c['SUB'])
lex = LVG(var_c)
query = GoogleQuery(lex)
assert query.startswith(
'"SCN5A" ("4786T>A"|"T4786A"|"4786T-->A"|"4786T/A"|"4786T->A"')
assert query.find('4732T->A') > -1
assert query.find('4783T-->A') > -1
def components_or_None(hgvs_p):
try:
comp = VariantComponents(Variant(hgvs_p))
if comp.ref != '':
return comp
except (TypeError, RejectedSeqVar, CriticalHgvsError):
# either the hgvs_p did not parse (Variant returned None) or it has incomplete edit info.
pass
return None
def hgvs2pmid_cli():
args = docopt(__doc__, version=__version__)
var = Variant(args['<hgvs>'])
if var:
hgvs2pmid(str(var))
else:
print(
'Supplied argument must be a valid HGVS string! See --help for examples.'
)
def test_crazy_long_hgvs_strings_just_to_see_what_happens(self):
# https://www.ncbi.nlm.nih.gov/clinvar/38771295/ RCV000008537
seqvar = Variant(
'NP_001121636.1:p.Gln197_Gln208delinsGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGlnGln'
) # https://www.ncbi.nlm.nih.gov/clinvar/36484151/
# https://www.ncbi.nlm.nih.gov/clinvar/variation/68075/
seqvar = Variant(
'LRG_702p1:p.Glu129_Asp145delinsGluIleLysValIleSerGlyIleLeuThrGlnGlyArgCysAspIleGluIleLysValIleSerGlyIleLeuThrGlnGlyArgCysAspIleAsp'
)
# RCV000087437
seqvar = Variant(
'LRG_3p1:p.Arg1139_Gly1140insValSerSerThrGluArgTyrTyrArgSerThrCysPheArgCysLeuHisPheArgLysIlePheTrpHisCysAspValMetIleLeuSerLeu'
)
# also RCV000087437
seqvar = Variant(
'NP_000081.1:p.Arg1139_Gly1140insValSerSerThrGluArgTyrTyrArgSerThrCysPheArgCysLeuHisPheArgLysIlePheTrpHisCysAspValMetIleLeuSerLeu'
)
def __init__(self, lex=None, seqvar=None, hgvs_text=None, **kwargs):
""" Requires either an LVG object (lex=) or a Sequence Variant object (seqvar=) or an hgvs_text string (hgvs_text=)
Priority for instantiation (in case of multiple-parameter submission): lex, seqvar, hgvs_text
Keywords:
gene_name: should be supplied when instantiated with seqvar= or hgvs_text=
"""
if lex:
self.lex = lex
self.seqvar = lex.seqvar
self.hgvs_text = lex.hgvs_text
if lex.gene_name:
self.gene_name = lex.gene_name
else:
self.gene_name = kwargs.get('gene_name', None)
elif seqvar:
self.lex = None
self.seqvar = seqvar
self.hgvs_text = '%s' % seqvar
self.gene_name = kwargs.get('gene_name', None)
elif hgvs_text:
self.lex = None
self.seqvar = Variant(hgvs_text)
self.hgvs_text = hgvs_text
self.gene_name = kwargs.get('gene_name', None)
if self.gene_name is None:
raise GoogleQueryMissingGeneName(
'Information supplied with variant %s is missing gene name.' %
self.seqvar)
# self.synonyms = {'c': [], 'g': [], 'p': [], 'n': []}
self.gene_synonyms = filter_gene_synonyms(GeneSynonyms(self.gene_name))
# choice of Google CSE ("cx") -- "whitelist" or "schema" [default: whitelist]
self.cse = kwargs.get('cse', 'whitelist')
def pubtator_results_for_seqvar(seqvar_or_hgvs_text, gene_id):
""" Takes a SequenceVariant or hgvs_text string.
Returns a dictionary of results mapping hgvs_text to a list of results from pubtator, i.e.:
{ hgvs_text: [ <dictionaries representing matching results from pubtator> ] }
:param seqvar_or_hgvs_text: hgvs_text or SequenceVariant object
:param gene_id: id of gene associated with variant (required)
:return: dictionary of results
:raises: RejectedSeqVar, PubtatorDBError
"""
seqvar = Variant(seqvar_or_hgvs_text)
hgvs_text = '%s' % seqvar
result = {hgvs_text: []}
components = VariantComponents(seqvar)
if seqvar.type == 'p':
result[hgvs_text] = pubtator_db.search_proteins(components, gene_id)
else:
result[hgvs_text] = pubtator_db.search_m2p(components, gene_id)
return result
def process_row(db, dbrow):
""" For each column in the row that might contain an HGVS string -- namely:
* variant_name
* HGVS_p
* HGVS_c
...do the following actions:
* if variant_name or HGVS_c, make an LVG. If both exist, make an lvg from HGVS_c (preferentially).
* for each seqvar, add a row to the database with '%s' % seqvar, PMID, VariantComponents --> Ref,Alt,Pos
* if HGVS_p, see if it exists in the LVG object (in hgvs_p).
* if not: try to make a VariantComponents object. add new row if comp is not None.
:param dbrow: (dict) one row from t2g_variant_summary table.
:return: list of VariantComponents successfully added to the database (or empty list)
"""
added_components = []
variants = {'c': [], 'g': [], 'n': []}
for option in ['variant_name']: #, 'HGVS_c']:
seqvar = seqvar_or_None(dbrow[option])
if seqvar:
variants[seqvar.type].append(seqvar)
# use the first usable variant to construct an LVG for this variant.
lex = None
for seqtype in ['c', 'g', 'n']:
if variants[seqtype]:
seqvar = variants[seqtype][0]
print('[%s] Using %s to build LVG' %
(dbrow['variant_name'], seqvar))
lex = lvg_or_None(seqvar)
if lex:
break
# collect a unique set of protein variants stripped of parentheses.
hgvs_ps = set()
#comp = components_or_None(dbrow['HGVS_p'])
#if comp:
# treat "uncertain" protein effects as unneeded duplicates of the "certain" ones.
# hgvs_ps.add(('%s' % Variant(dbrow['HGVS_p'])).replace(')','').replace('(',''))
# add non-p-vars to the database
if lex:
for item in lex.hgvs_p:
hgvs_ps.add(item.replace(')', '').replace('(', ''))
for seqtype in ['c', 'g', 'n']:
for seqvar in lex.variants[seqtype].values():
comp = components_or_None(seqvar)
if comp:
if add_components_to_row(db, dbrow, comp):
added_components.append(comp)
# and now add the proteins
for hgvs_p in hgvs_ps:
comp = components_or_None(Variant(hgvs_p))
if comp:
if add_components_to_row(db, dbrow, comp):
added_components.append(comp)
return added_components
def test_frameshift_case(self):
var_c = Variant(hgvs_c['FS'])
lex = LVG(var_c)
query = GoogleQuery(lex)
def test_duplication(self):
var_c = Variant(hgvs_c['DUP'])
comp = VariantComponents(var_c)
pass
def test_insert(self):
var_c = Variant(hgvs_c['INS'])
comp = VariantComponents(var_c)
pass
def test_frameshift(self):
var_p = Variant(hgvs_p['FS'])
comp = VariantComponents(var_p)
pass
def test_indel(self):
var_n = Variant(hgvs_n['INDEL'])
comp = VariantComponents(var_n)
pass
def test_simple_substitution(self):
var_c = Variant(hgvs_c['SUB'])
comp = VariantComponents(var_c)
pass
def test_deletion(self):
var_g = Variant(hgvs_g['DEL'])
comp = VariantComponents(var_g)
pass
def test_deletion_case(self):
var_c = Variant(hgvs_c['DEL'])
lex = LVG(var_c)
query = GoogleQuery(lex)
def test_dup_case(self):
var_c = Variant(hgvs_c['DUP'])
lex = LVG(var_c)
query = GoogleQuery(lex)
def test_ins_case(self):
var_c = Variant(hgvs_c['INS'])
lex = LVG(var_c)
query = GoogleQuery(lex)
def test_allow_gene_name_in_hgvs_string(self):
seqvar = Variant('NM_001165963.1(SCN1A):c.4338_6030del')
assert 'NM_001165963.1:c.4338_6030del' == '%s' % seqvar
def test_bad_hgvs_string_returns_None(self):
assert Variant('boogers') is None
