def testManyPunctations(self):
text = u'In ' \
u'addition, the antagonistic action of propranolol (1 X 10(-7) M) in a Ca++-containing or ' \
u'Sr++-containing medium was determined. '
sentence = SentenceAnalysisSpacy(text, self.nlp)
sentence.analyse()
def test_asthma(self):
text = u'Asthma is a chronic disease characterized by airway inflammation, obstruction and hyperresponsiveness.'
expected_noun_phrases = set([
'chronic disease', 'airway inflammation', 'obstruction', 'Asthma',
'hyperresponsiveness'
])
sentence = SentenceAnalysisSpacy(text, self.nlp)
sentence.analyse()
noun_phrases = set([i.text for i in sentence.noun_phrases])
self.assertTrue(
_concept_exists(subject=u'Asthma',
verb=u'be',
object=u'chronic disease',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'Asthma',
verb=u'be characterized by',
object=u'hyperresponsiveness',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'Asthma',
verb=u'be characterized by',
object=u'airway inflammation',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'Asthma',
verb=u'be characterized by',
object=u'obstruction',
concept_list=sentence.concepts))
self.assertEqual(noun_phrases, expected_noun_phrases)
def alpha_syn(self):
text = u'Deubiquitinase Usp8 regulates α-synuclein clearance and modifies its toxicity in Lewy body disease.'
expected_noun_phrases = set([
'Usp8', 'Lewy body disease', 'alpha-synuclein clearance',
'toxicity'
])
sentence = SentenceAnalysisSpacy(text, self.nlp)
sentence.analyse()
noun_phrases = set([i.text for i in sentence.noun_phrases])
self.assertEqual(noun_phrases, expected_noun_phrases)
self.assertTrue(
_concept_exists(subject=u'Usp8',
verb=u'regulate',
object=u'alpha-synuclein clearance',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'Usp8',
verb=u'regulate modifies',
object=u'Lewy body disease',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'Usp8',
verb=u'regulate modifies',
object=u'toxicity',
concept_list=sentence.concepts))
def test_multi_gene_and_disease(self):
text = u' Molecular genetics and developmental studies have identified 21 genes in this region (ADRA1A, ' \
u'ARHGEF10, CHRNA2, CHRNA6, CHRNB3, DKK4, DPYSL2, EGR3, FGF17, FGF20, ' \
u'FGFR1, FZD3, LDL, NAT2, NEF3, NRG1, PCM1, PLAT, ' \
u'PPP3CC, SFRP1 and VMAT1/SLC18A1) that are most likely to contribute to neuropsychiatric disorders ' \
u'(schizophrenia, autism, bipolar disorder and depression), neurodegenerative disorders (Parkinson\'s' \
u' and Alzheimer\'s disease) and cancer.'
minimal_expected_noun_phrases = [
'autism', 'ARHGEF10', 'NEF3', 'genes', 'depression', 'CHRNA6',
'PCM1', 'DKK4', 'PPP3CC', 'EGR3', 'VMAT1/SLC18A1', 'FGF20',
'bipolar disorder', 'CHRNA2', 'FZD3', 'Molecular genetics',
'CHRNB3', 'NAT2', 'DPYSL2', 'NRG1', 'cancer', 'FGF17', 'PLAT',
'FGFR1', 'SFRP1', 'neuropsychiatric disorders', 'region', 'LDL',
'schizophrenia', 'depression', 'Parkinson\'s',
'Alzheimer\'s disease'
]
sentence = SentenceAnalysisSpacy(text, self.nlp)
sentence.analyse()
noun_phrases = set([i.text for i in sentence.noun_phrases])
self.assertTrue(
_concept_exists(subject=u' Molecular genetics',
verb=u'identify',
object=u'FZD3',
concept_list=sentence.concepts))
for i in minimal_expected_noun_phrases:
self.assertIn(i, noun_phrases)
def test_Schistosoma(self):
text = u'Studies have suggested that Schistosoma mansoni infection reduces the severity of asthma and prevent ' \
u'' \
u'' \
u'' \
u'' \
u'' \
u'atopy.'
sentence = SentenceAnalysisSpacy(text, self.nlp)
sentence.analyse()
noun_phrases = set([i.text for i in sentence.noun_phrases])
self.assertIn(u'Schistosoma mansoni infection', noun_phrases)
self.assertTrue(
_concept_exists(subject=u'Schistosoma mansoni infection',
verb=u'suggest reduces',
object=u'asthma',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'Schistosoma mansoni infection',
verb=u'suggest prevent',
object=u'atopy',
concept_list=sentence.concepts))
def test_hyphen_token(self):
text = u'Here we report that the Polo-like kinase PLK1, an essential mitotic kinase regulator, ' \
u'is an important downstream effector of c-ABL in regulating the growth of cervical cancer.'
sentence = SentenceAnalysisSpacy(text, self.nlp)
sentence.analyse()
noun_phrases = set([i.text for i in sentence.noun_phrases])
self.assertIn(u'Polo-like kinase PLK1', noun_phrases)
self.assertIn(u'c-ABL', noun_phrases)
self.assertTrue(
_concept_exists(subject=u'Polo-like kinase PLK1',
verb=u'report is',
object=u'important downstream effector',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'Polo-like kinase PLK1',
verb=u'report is',
object=u'c-ABL',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'Polo-like kinase PLK1',
verb=u'report regulating',
object=u'cervical cancer',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'Polo-like kinase PLK1',
verb=u'report regulating',
object=u'growth',
concept_list=sentence.concepts))
def test_clinical_trials_and_il5_antiodies(self):
text = u'Recently, more and more clinical trials have been performed to evaluate the effects of ' \
u'anti-interleukin (IL)-5 antibodies in eosinophilic asthma.'
# TODO: should be this:
# expected_noun_phrases = set(
# ['anti-interleukin (IL)-5 antibodies', 'effects', 'clinical trials', 'eosinophilic asthma'])
expected_noun_phrases = set([
'anti-interleukin', 'effects', 'clinical trials',
'eosinophilic asthma'
])
sentence = SentenceAnalysisSpacy(text, self.nlp)
sentence.analyse()
noun_phrases = set([i.text for i in sentence.noun_phrases])
self.assertTrue(
_concept_exists(subject=u'clinical trials',
verb=u'perform evaluate',
object=u'effects',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'clinical trials',
verb=u'perform evaluate',
object=u'eosinophilic asthma',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'clinical trials',
verb=u'perform evaluate',
object=u'anti-interleukin',
concept_list=sentence.concepts))
self.assertEqual(noun_phrases, expected_noun_phrases)
def test_doc(self):
text = u'Asthma is a chronic disease characterized by airway inflammation, obstruction and hyperresponsiveness.'
doc = self.nlp(text)
sentence = SentenceAnalysisSpacy(doc, self.nlp)
sentence.analyse()
self.assertTrue(
_concept_exists(subject=u'Asthma',
verb=u'be',
object=u'chronic disease',
concept_list=sentence.concepts))
def test_span(self):
text = u'Asthma is a chronic disease characterized by airway inflammation, obstruction and ' \
u'hyperresponsiveness. ' \
u'Severe asthma affects a small proportion of subjects but results in most of the morbidity, ' \
u'costs and mortality ' \
u'associated with the disease.'
doc = self.nlp(text)
for span in doc.sents:
sentence = SentenceAnalysisSpacy(span, self.nlp)
sentence.analyse()
self.assertTrue(
_concept_exists(subject=u'Asthma',
verb=u'be',
object=u'chronic disease',
concept_list=sentence.concepts))
break
def test_to_text(self):
text = u'Molecular genetics and developmental studies have identified 21 genes in this region (ADRA1A, ' \
u'ARHGEF10, CHRNA2, CHRNA6, CHRNB3, DKK4, DPYSL2, EGR3, FGF17, FGF20, ' \
u'FGFR1, FZD3, LDL, NAT2, NEF3, NRG1, PCM1, PLAT, ' \
u'PPP3CC, SFRP1 and VMAT1/SLC18A1) that are most likely to contribute to neuropsychiatric disorders ' \
u'(schizophrenia, autism, bipolar disorder and depression), neurodegenerative disorders (Parkinson\'s' \
u' and Alzheimer\'s disease) and cancer.'
sentence = SentenceAnalysisSpacy(text, self.nlp)
sentence.analyse(verbose=True, )
print sentence.to_text()
print sentence.to_pos_tagged_text()
def test_serum_level(self):
'''test verb descriptor to be collected'''
text = u'The serum levels of CA125, CA15.3, and HE4 were significantly higher in the TTF-1-positive group ' \
u'than in the TTF-1-negative group (p<0.05).'
expected_noun_phrases = set(
['TTF-1-negative group', 'serum levels', 'TTF-1-positive group'])
sentence = SentenceAnalysisSpacy(text, self.nlp)
sentence.analyse()
noun_phrases = set([i.text for i in sentence.noun_phrases])
self.assertEqual(noun_phrases, expected_noun_phrases)
self.assertTrue(
_concept_exists(subject=u'serum levels',
verb=u'be higher',
object=u'TTF-1-positive group',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'serum levels',
verb=u'be higher than',
object=u'TTF-1-negative group',
concept_list=sentence.concepts))
def test_Fanconi(self):
text = u'Fanconi anemia (FA) is a genetic disease characterized by bone marrow failure and increased cancer ' \
u'risk.'
expected_noun_phrases = set([
'bone marrow failure',
'Fanconi anemia',
'cancer risk',
'genetic disease',
])
sentence = SentenceAnalysisSpacy(text, self.nlp)
self.assertIn(u'FA', sentence.abbreviations)
self.assertEqual(u'Fanconi anemia', sentence.abbreviations[u'FA'])
sentence.analyse()
noun_phrases = set([i.text for i in sentence.noun_phrases])
self.assertEqual(noun_phrases, expected_noun_phrases)
self.assertTrue(
_concept_exists(subject=u'Fanconi anemia',
verb=u'be characterized by',
object=u'cancer risk',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'genetic disease',
verb=u'be characterized by',
object=u'cancer risk',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'Fanconi anemia',
verb=u'be characterized by',
object=u'bone marrow failure',
concept_list=sentence.concepts))
self.assertTrue(
_concept_exists(subject=u'genetic disease',
verb=u'be characterized by',
object=u'bone marrow failure',
concept_list=sentence.concepts))
def test_custom_tokenizer(self):
text = u'the antagonistic action of propranolol (1 X 10(-7) M) in a Ca++-containing or. Cell growth and ' \
u'quabain-sensitive 86Rg+ uptake and (Na++K+)-ATPase activity in 3T3 and SV40 transformed 3T3 ' \
u'fibroblasts. The uptake of ouabain-sensitive 86Rb+ uptake measured at 5 min and the uptake measured ' \
u'at 60 min was 4.5- and 2.7-fold greater respectively for SV40 transformed 3T3 cells compared to 3T3 ' \
u'cells during the late log phase of growth. This uptake, however, varied markedly with cell growth. ' \
u'Ouabain-sensitive 86Rb+ uptake was found to be a sensitive indicator of protein synthesis as ' \
u'measured by total protein content. Cessation of cell growth as measured by total protein content was ' \
u'' \
u'' \
u'' \
u'associated with a decline in ouabain-sensitive 86Rb+ uptake in both cell types. This increase ' \
u'ouabain-sensitive cation transport was reflected in increased levels of (Na++K)-ATPase activity for ' \
u'SV40 3T3 cells, which showed a 2.5-fold increase V but the same Km as 3T3 cells. These results are ' \
u'compared with the results of related work. Possible mechanisms for these effects are discussed and ' \
u'how changes in cation transport might be related to alterations in cell growth. This is a test, ' \
u'for a complex entity name: th:is.{e}nt/ity-is,ver-y/co_m[p]lex(to)par;se . '
# u'Derivatives of 1,2,3,11a-tetrahydro-5H-pyrrolo[2,1-c][1,4]benzodiazepine-5,11(10H)-dione as ' \
# u'anxiolytic agents. A study of the pharmacological properties of pyrrolo[2,1-c][1,4]benzodiazepine ' \
# u'derivatives led to the choice of (+)-1,2,3,11a-tetrahydro-10-methyl-5H-pyrrolol[2,1-c][1,' \
# u'4]benzodiazepine-5,11)10H)-dione as a candidate for anxiolytic evaluation in a limited clinical ' \
# u'trial in man. Metabolism studies in laboratory animals have pointed to rapid hydroxylation, ' \
# u'possibly in the 3 and 11a positions. A series of compouds containing methyl groups in one or more of ' \
# u'these positions has been prepared in an effort to block metabolism and thereby obtain more active or ' \
# u'longer acting compounds. All of these derivatives were less active than the parent compound.'
# u'Inversion of optical configuration of alpha-methylfluorene-2-acetic acid (cicloprofen) in rats and
# monkeys. A simple and sensitive radiometric method to determine the individual enantiomers of cicloprofen
# has been developed. 14C-Cicloprofen was converted to its L-leucine diastereoisomers, which were separated
# by thin-layer chromatography and quantified by measuring the radioactivity in the area corresponding to
# each individual diastereoisomer. This technique has also been used to measure the enantiomers of unlabeled
# cicloprofen by condensing with 14C-labeled L-leucine. By using the radiometric method,
# a unique biotransformation process, the inversion of the (-)-enantiomer of alpha-methylfluorene-2-acetic
# acid to its (+)-enantiomer, has been demonstrated in the rat and monkey. The rate of (-)- to (+)-inversion
# was found to be faster in the rat than in the monkey. After single or repeated oral adminstration of the
# racemic modification or the (-)-enantiomer of cicloprofen to both species, the ratio of (+)- to (
# -)-enantiomers of cicloprofen in plasma, urine, or bile increased with time. At 5, 22, and 48 hr after oral
# administration of a single 50-mg/kg dose of the (-)-enantiomer, 14C-cicloprofen in rat plasma contained
# 20, 50, and 79%, respectively, of the (+)-enantiomer. After receiving the same dose of (-)-enantiomer,
# monkey plasma contained 16.5% and 32% of (+)-enantiomer at 8 and 24 hr, respectively. After oral
# administration of a single 50-mg/kg dose of the (+)-enantiomer of 14C-cicloprofen to rats and monkeys,
# the percentage of (-)-enantiomer in plasma varied from 2 to 15%. Since the administered (+)-enantiomer
# contained 4% of (-)-enantiomer and the (+)-enantiomer was excreted at a faster rate than its (-)-antipode
# by rats or monkeys, it is not known whether an occasional small percentage increase of (-)-enantiomer in
# plasma resulted from the (+)-to-(-) inversion, or from faster elimination of the (+)-enantiomer.
# Nevertheless, if (+)-to-(-) inversion does occur in these two species, the rate is much slower than for the
# (-)-to-(+) inversion.'
# u' Properties of common wheat ferredoxin, and a comparison with ferredoxins from related species of
# triticum and aegilops. Wheat ferredoxin was purified from the leaves of common wheat (Triticum aestivum).
# The absorption spectrum showed maxima at 465, 425, 332, and 278 nm. The absorbance ratio, A425 nm/A278 nm
# was 0.49, and the millimolar extinction coefficient at 425 nm was 10.8 mM-1. cm-1. The amino acid
# composition was determined to be Lys5, His2, Arg1, Asp11, Thr5, Ser7, Glu18, Pro5, Gly6, Ala7, Cys5, Val7,
# Met1, Ile4, Leu7, Tyr4, Phe1, and Trp1. The total number of amino acid residues was 97. The molecular
# weight was calculated from the amino acid composition to be 10,829, including iron and sulfur atoms. This
# value was confirmed by other methods, which were based on the contents of non-heme iron and of terminal
# amino acid. The N-terminal amino acid was alanine, and the C-terminal amino acid sequence was
# -Glu-Leu-Thr-AlaCOOH. Comparative studies were performed between T. aestivum ferredoxin and ferredoxins
# isolated from closely related species; these were T. aegilopoides, T. durum, Ae. squarrosa, and Ae. ovata.
# No significant differences in the properties of these ferredoxins were detected. It was also shown that
# these ferredoxins are immunologically homologous. It is, therefore, likely that one molecular species of
# ferredoxin is distributed through two genera of Triticum and Aegilops.'
sentence = SentenceAnalysisSpacy(text, self.nlp)
sentence.analyse(merge_with_syntax=False)
tokens = [i.text for i in sentence.doc]
self.assertIn(u'10(-7)', tokens)
self.assertIn(u'(Na++K+)-ATPase', tokens)
self.assertIn(u'2.7-fold', tokens)
self.assertIn(u'4.5-', tokens)
self.assertIn(u'86Rb+', tokens)
self.assertIn(u'Ca++-containing', tokens)
self.assertIn(u'(Na++K)-ATPase', tokens)
self.assertIn(u'Ouabain-sensitive', tokens)
self.assertIn(u'th:is.{e}nt/ity-is,ver-y/co_m[p]lex(to)par;se', tokens)
self.assertNotIn(u'cells,', tokens)
self.assertNotIn(u'(1', tokens)
self.assertNotIn(u'fibroblasts.', tokens)