From 636e2166164a56cf26361dd4369526c2ac45663d Mon Sep 17 00:00:00 2001 From: Hao Chen Date: Mon, 6 May 2019 10:45:20 -0500 Subject: three types of searches --- .gitignore | 1 + get_addiction_sentences.py | 92 ++++++++++++++++++++++++++++++---------------- 2 files changed, 61 insertions(+), 32 deletions(-) create mode 100644 .gitignore mode change 100644 => 100755 get_addiction_sentences.py diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..355a0c3 --- /dev/null +++ b/.gitignore @@ -0,0 +1 @@ +.key diff --git a/get_addiction_sentences.py b/get_addiction_sentences.py old mode 100644 new mode 100755 index 30a8e50..e24f122 --- a/get_addiction_sentences.py +++ b/get_addiction_sentences.py @@ -7,29 +7,9 @@ import sys gene=sys.argv[1] -addiction_terms="sensitization|intake|addiction|drug abuse|relapse|self-administered|self-administration|voluntary|reinstatement|binge|intoxication|withdrawal|chronic" - -drugs="alcohol|alcoholism|smoking|nicotine|tobacco|methamphetamine|amphetamine|cocaine|opioid|fentanyl|oxycodone|oxycontin|heroin|morphine|marijuana|cannabinoid|tetrahydrocannabinol|thc" - -brain_regions="cortex|accumbens|striatum|amygadala|hippocampus|tegmental|mesolimbic|infralimbic|prelimbic" - -brain_d ={"cortex":"cortex|pfc|vmpfc|il|pl|prelimbic|infralimbic", - "striatum":"striatum|STR", - "accumbens":"shell|core|NAcc|acbs|acbc", - "hippocampus":"hippocampus|hipp|hip|ca1|ca3|dentate|gyrus", - "amygadala":"amygadala|cea|bla|amy", - "ventral tegmental":"ventral tegmental|vta" - } - -function="LTP|LTD|plasticity|regulate|glutamate|GABA|cholinergic|serotoninergic|synaptic|methylation|transcription|phosphorylation" - -drugs_d = {"alcohol":"alcohol|alcoholism", - "nicotine":"smoking|nicotine|tobacco", - "amphetamine":"methamphetamine|amphetamine", - "cocaine":"cocaine", - "opioid":"opioid|fentanyl|oxycodone|oxycontin|heroin|morphine", - "cannabinoid":"marijuana|cannabinoid|Tetrahydrocannabinol|thc" - } +## turn dictionary (synonyms) to regular expression +def undic(dic): + return "|".join(dic.keys())+"|"+"|".join(dic.values()) def findWholeWord(w): return re.compile(r'\b({0})\b'.format(w), flags=re.IGNORECASE).search @@ -42,6 +22,7 @@ def getSentences(query): pmid = tiab.pop(0) tiab= " ".join(tiab) sentences = sent_tokenize(tiab) + ## keep the sentence only if it contains the gene for sent in sentences: if findWholeWord(gene)(sent): sent=re.sub(r'\b(%s)\b' % gene, r'\1', sent, flags=re.I) @@ -49,32 +30,79 @@ def getSentences(query): return(out) def gene_addiction(gene): - q="\"(" + addiction_terms.replace("|", " OR ") + ") AND (" + drugs.replace("|", " OR ", ) + ") AND " + gene + "\"" + # search gene name & drug name in the context of addiction terms (i.e., exclude etoh affects cancer, or methods to extract cocaine) + q="\"(" + addiction.replace("|", " OR ") + ") AND (" + drugs.replace("|", " OR ", ) + ") AND " + gene + "\"" sents=getSentences(q) out=str() for sent in sents.split("\n"): for drug0 in drugs_d: if findWholeWord(drugs_d[drug0])(sent) : sent=re.sub(r'\b(%s)\b' % drugs_d[drug0], r'\1', sent, flags=re.I) - out+=gene+"\t"+drug0+"\t"+sent+"\n" + out+=gene+"\t"+"drug\t" + drug0+"\t"+sent+"\n" + for add0 in addiction.split("|"): + if findWholeWord(add0)(sent) : + sent=re.sub(r'\b(%s)\b' % add0, r'\1', sent, flags=re.I) + out+=gene+"\t"+"addiction\t"+add0+"\t"+sent+"\n" return(out) -def gene_brainRegion(gene): - q="\"(" + brain_regions.replace("|", " OR ") + ") AND " + gene + "\"" +def gene_anatomical(gene): + q="\"(" + brain.replace("|", " OR ") + ") AND " + gene + "\"" sents=getSentences(q) out=str() for sent in sents.split("\n"): for brain0 in brain_d: if findWholeWord(brain_d[brain0])(sent) : sent=re.sub(r'\b(%s)\b' % brain_d[brain0], r'\1', sent, flags=re.I) - out+=gene+"\t"+brain0+"\t"+sent+"\n" + out+=gene+"\t"+"brain\t"+brain0+"\t"+sent+"\n" + return(out) + +def gene_biological(gene): + q="\"(" + biological.replace("|", " OR ") + ") AND " + gene + "\"" + sents=getSentences(q) + out=str() + for sent in sents.split("\n"): + for bio0 in biological_d: + if findWholeWord(biological_d[bio0])(sent) : + sent=re.sub(r'\b(%s)\b' % biological_d[bio0], r'\1', sent, flags=re.I) + out+=gene+"\t"+"function\t"+bio0+"\t"+sent+"\n" return(out) +addiction="reward|reinforcement|sensitization|intake|addiction|drug abuse|relapse|self-administered|self-administration|reinstatement|binge|intoxication|withdrawal|conditioned place preference|aversion|aversive|CPP" + +drugs_d = {"alcohol":"alcohol|alcoholism", + "nicotine":"smoking|nicotine|tobacco", + "amphetamine":"methamphetamine|amphetamine", + "cocaine":"cocaine", + "opioid":"opioid|fentanyl|oxycodone|oxycontin|heroin|morphine", + "cannabinoid":"marijuana|cannabinoid|Tetrahydrocannabinol|thc" + } +drugs=undic(drugs_d) + +brain_d ={"cortex":"cortex|pfc|vmpfc|il|pl|prelimbic|infralimbic", + "striatum":"striatum|STR", + "accumbens":"shell|core|NAcc|acbs|acbc", + "hippocampus":"hippocampus|hipp|hip|ca1|ca3|dentate|gyrus", + "amygadala":"amygadala|cea|bla|amy", + "vta":"ventral tegmental|vta|pvta" + } +# brain region has too many short acronyms to just use the undic function, so search PubMed using the following +brain="cortex|accumbens|striatum|amygadala|hippocampus|tegmental|mesolimbic|infralimbic|prelimbic" + +biological_d={"plasticity":"LTP|LTD|plasticity|synaptic|epsp|epsc", + "neurotransmission": "neurotransmission|glutamate|GABA|cholinergic|serotoninergic", + "signalling":"signalling|phosphorylation|glycosylation", +# "regulation":"increased|decreased|regulated|inhibited|stimulated", + "transcription":"transcription|methylation|histone|ribosome", + } +biological=undic(biological_d) + report=str() -out=gene_addiction(gene) -report+=out -out=gene_brainRegion(gene) -report+=out +out0=gene_addiction(gene) +report+=out0 +out1=gene_anatomical(gene) +report+=out1 +out2=gene_biological(gene) +report+=out2 with codecs.open(gene+"_addiction_sentences.tab", "w", encoding='utf8') as writer: writer.write(report) writer.close() -- cgit v1.2.3