diff options
| -rw-r--r-- | .gitignore | 1 | ||||
| -rwxr-xr-x | gatpub.py | 133 | ||||
| -rw-r--r-- | get_addiction_sentences.py | 83 | ||||
| -rw-r--r-- | server.py | 32 |
4 files changed, 166 insertions, 83 deletions
diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..355a0c3 --- /dev/null +++ b/.gitignore @@ -0,0 +1 @@ +.key 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'<b>\1</b>', sent, flags=re.I) + out+=pmid+"\t"+sent+"<br>\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'<b>\1</b>', 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'<b>\1</b>', 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'<b>\1</b>', 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'<b>\1</b>', 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 diff --git a/get_addiction_sentences.py b/get_addiction_sentences.py deleted file mode 100644 index 30a8e50..0000000 --- a/get_addiction_sentences.py +++ /dev/null @@ -1,83 +0,0 @@ -#!/bin/env python3 -from nltk.tokenize import sent_tokenize -import os -import re -import codecs -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" - } - -def findWholeWord(w): - return re.compile(r'\b({0})\b'.format(w), flags=re.IGNORECASE).search - -def getSentences(query): - 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) - for sent in sentences: - if findWholeWord(gene)(sent): - sent=re.sub(r'\b(%s)\b' % gene, r'<b>\1</b>', sent, flags=re.I) - out+=pmid+"\t"+sent+"\n" - return(out) - -def gene_addiction(gene): - q="\"(" + addiction_terms.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'<b>\1</b>', sent, flags=re.I) - out+=gene+"\t"+drug0+"\t"+sent+"\n" - return(out) - -def gene_brainRegion(gene): - q="\"(" + brain_regions.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'<b>\1</b>', sent, flags=re.I) - out+=gene+"\t"+brain0+"\t"+sent+"\n" - return(out) - -report=str() -out=gene_addiction(gene) -report+=out -out=gene_brainRegion(gene) -report+=out -with codecs.open(gene+"_addiction_sentences.tab", "w", encoding='utf8') as writer: - writer.write(report) - writer.close() - - - diff --git a/server.py b/server.py new file mode 100644 index 0000000..907877a --- /dev/null +++ b/server.py @@ -0,0 +1,32 @@ +from flask import Flask, render_template, request, redirect +import simplejson as json +from gatpub import * + +app=Flask(__name__) +app.config['SECRET_KEY'] = '#DtfrL98G5t1dC*4' + +@app.route("/") +def root(): + return render_template('index.html') + +@app.route("/home") +def home(): + return render_template('index.html') + +@app.route("/network", methods=['GET', 'POST']) +def network(): + edges_list=[] + nodes_list=[] + if request.method == 'POST': + term = request.form + gene=term['query'] + nodes="{ data: { id: '" + gene + "'} },\n" + default_nodes + tmp0=gene_addiction(gene) + e0=generate_edges(tmp0) + tmp1=gene_functional(gene) + e1=generate_edges(tmp1) + tmp2=gene_anatomical(gene) + e2=generate_edges(tmp2) + return render_template('network.html', elements=nodes+e0+e1+e2) +if __name__ == '__main__': + app.run(debug=True) |