From f09a791cf83f71b930faf842d7d5806627d6aa84 Mon Sep 17 00:00:00 2001
From: Hao Chen
Date: Tue, 7 May 2019 16:30:29 -0500
Subject: basic cytoscape plot
---
gatpub.py | 133 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 133 insertions(+)
create mode 100755 gatpub.py
(limited to 'gatpub.py')
diff --git a/gatpub.py b/gatpub.py
new file mode 100755
index 0000000..6647e32
--- /dev/null
+++ b/gatpub.py
@@ -0,0 +1,133 @@
+#!/bin/env python3
+from nltk.tokenize import sent_tokenize
+import os
+import re
+import codecs
+import sys
+
+#gene=sys.argv[1]
+
+## turn dictionary (synonyms) to regular expression
+def undic(dic):
+ return "|".join(dic.values())
+
+def findWholeWord(w):
+ return re.compile(r'\b({0})\b'.format(w), flags=re.IGNORECASE).search
+
+def getSentences(query, gene):
+ abstracts = os.popen("esearch -db pubmed -query " + query + " | efetch -format uid |fetch-pubmed -path /run/media/hao/PubMed/Archive/ | xtract -pattern PubmedArticle -element MedlineCitation/PMID,ArticleTitle,AbstractText").read()
+ out=str()
+ for row in abstracts.split("\n"):
+ tiab=row.split("\t")
+ 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)
+ out+=pmid+"\t"+sent+"
\n"
+ return(out)
+
+def gene_addiction(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 (" + drug.replace("|", " OR ", ) + ") AND " + gene + "\""
+ sents=getSentences(q, gene)
+ out=str()
+ for sent in sents.split("\n"):
+ for drug0 in drug_d:
+ if findWholeWord(drug_d[drug0])(sent) :
+ sent=re.sub(r'\b(%s)\b' % drug_d[drug0], r'\1', sent, flags=re.I)
+ 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_anatomical(gene):
+ q="\"(" + brain.replace("|", " OR ") + ") AND " + gene + "\""
+ sents=getSentences(q,gene)
+ 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"+"brain\t"+brain0+"\t"+sent+"\n"
+ return(out)
+
+def gene_functional(gene):
+ q="\"(" + function.replace("|", " OR ") + ") AND " + gene + "\""
+ sents=getSentences(q,gene)
+ out=str()
+ for sent in sents.split("\n"):
+ for bio0 in function_d:
+ if findWholeWord(function_d[bio0])(sent) :
+ sent=re.sub(r'\b(%s)\b' % function_d[bio0], r'\1', sent, flags=re.I)
+ out+=gene+"\t"+"function\t"+bio0+"\t"+sent+"\n"
+ return(out)
+
+def generate_nodes(nodes_d):
+ # include all search terms even if there are no edges, just to show negative result
+ json0 =str() #"{ data: { id: '" + gene + "'} },\n"
+ for node in nodes_d:
+ json0 += "{ data: { id: '" + node + "'} },\n"
+ return(json0)
+
+def generate_edges(data):
+ json0=str()
+ for line in data.split("\n"):
+ if len(line.strip())!=0:
+ (source, cat, target, pmid, sent) = line.split("\t")
+ edgeID=source+"_"+target
+ json0+="{ data: { id: \'" + edgeID + "\', source: \'" + source + "\', target: '" + target + "\' } },\n"
+ return(json0)
+
+addiction_d = {"reward":"reward|reinforcement|conditioned place preference|CPP|self-administration|self-administered",
+ "aversion":"aversion|aversive|conditioned taste aversion|CTA|withdrawal",
+ "relapse":"relapse|reinstatement|craving|drug seeking",
+ "sensitization":"sensitization",
+ "addiction":"addiction|drug abuse",
+ "intoxication":"intoxication|binge",
+ }
+addiction=undic(addiction_d)
+
+drug_d = {"alcohol":"alcohol|alcoholism",
+ "nicotine":"smoking|nicotine|tobacco",
+ "amphetamine":"methamphetamine|amphetamine|METH",
+ "cocaine":"cocaine",
+ "opioid":"opioid|fentanyl|oxycodone|oxycontin|heroin|morphine",
+ "cannabinoid":"marijuana|cannabinoid|tetrahydrocannabinol|thc|thc-9"
+ }
+drug=undic(drug_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"
+
+function_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",
+ }
+function=undic(function_d)
+
+#out1=gene_anatomical(gene)
+#out2=gene_functional(gene)
+#report=out0+out1+out2
+#with codecs.open(gene+"_addiction_sentences.tab", "w", encoding='utf8') as writer:
+# writer.write(report)
+# writer.close()
+
+n0=generate_nodes(function_d)
+n1=generate_nodes(addiction_d)
+n2=generate_nodes(drug_d)
+n3=generate_nodes(brain_d)
+default_nodes=n0+n1+n2+n3
--
cgit v1.2.3