From fab561751530ed01d7db843250d88941e3663152 Mon Sep 17 00:00:00 2001
From: Piotr Gawron <piotr.gawron@uni.lu>
Date: Mon, 31 Jul 2017 15:17:57 +0200
Subject: [PATCH] parsing of drugbank targets improved
when target didn't have references some targets might get skipped
---
.../services/DrugbankHTMLParser.java | 126 ++++++++++--------
.../services/DrugbankHTMLParserTest.java | 9 ++
.../testFiles/drugbank/target-html-part.html | 1 +
3 files changed, 81 insertions(+), 55 deletions(-)
create mode 100644 annotation/testFiles/drugbank/target-html-part.html
diff --git a/annotation/src/main/java/lcsb/mapviewer/annotation/services/DrugbankHTMLParser.java b/annotation/src/main/java/lcsb/mapviewer/annotation/services/DrugbankHTMLParser.java
index fcdf9c0225..537492673d 100644
--- a/annotation/src/main/java/lcsb/mapviewer/annotation/services/DrugbankHTMLParser.java
+++ b/annotation/src/main/java/lcsb/mapviewer/annotation/services/DrugbankHTMLParser.java
@@ -300,7 +300,7 @@ public class DrugbankHTMLParser extends DrugAnnotation implements IExternalServi
List<Target> getTargetsForDrug(String page) throws DrugSearchException {
List<Target> result = new ArrayList<>();
try {
- int i, j;
+ int i;
Target target = new Target();
target.setType(TargetType.SINGLE_PROTEIN);
@@ -316,63 +316,18 @@ public class DrugbankHTMLParser extends DrugAnnotation implements IExternalServi
i = page.indexOf("Details</a></div>", i);
while (i > 0 && i < end) {
- int kindIndex = page.indexOf("<dt>Kind</dt><dd>", i);
- kindIndex += "<dt>Kind</dt><dd>".length();
- int endKindIndex = page.indexOf("</dd>", kindIndex);
- String type = "";
- if (kindIndex > 0 && endKindIndex > kindIndex) {
- type = page.substring(kindIndex, endKindIndex);
+ int targetStart = page.indexOf("</div><strong>", i);
+ int nextTargetStart = page.indexOf("</div><strong>", targetStart + 1);
+ if (nextTargetStart < 0) {
+ nextTargetStart = end;
}
- if (type.trim().equalsIgnoreCase("Protein")) {
- i = page.indexOf("/biodb/polypeptides/", i);
- i += "/biodb/polypeptides/".length();
- target = new Target();
- target.setType(TargetType.SINGLE_PROTEIN);
-
- // Getting ID && Name
- j = page.indexOf('"', i);
- String uniprotId = page.substring(i, j);
- MiriamData uniprotTarget = new MiriamData(MiriamType.UNIPROT, uniprotId);
- MiriamData hgncTarget = uniprotAnnotator.uniProtToHgnc(uniprotTarget);
- if (hgncTarget != null) {
- target.addGene(hgncTarget);
- } else {
- target.addGene(uniprotTarget);
- }
-
- i = j + 2;
- j = page.indexOf("</", i);
- String name = StringEscapeUtils.unescapeHtml4(page.substring(i, j));
- target.setName(name);
-
- // Getting Organism
- i = page.indexOf("Organism</dt><dd>", i);
- i = i + "Organism</dt><dd>".length();
- j = page.indexOf("</dd>", i);
- target.setOrganism(getTaxonomyBackend().getByName(page.substring(i, j)));
-
- // Getting References
- i = page.indexOf("<strong>References</strong>", i);
- if (i > 0 && i < end) {
- int nextI = Math.min(page.indexOf("Details</a></div>", i), end);
- if (nextI < 0) {
- nextI = end;
- }
- target.addReferences(getPubmedFromRef(page.substring(i, nextI)));
- i = nextI;
- } else {
- i = end;
- }
+
+ target = parseTarget(page.substring(targetStart, nextTargetStart));
+ if (target != null) {
result.add(target);
- } else {
- i = kindIndex;
- logger.warn("Unknown target type: " + type + ". Skipping.");
- int nextI = Math.min(page.indexOf("Details</a></div>", i), end);
- if (nextI < 0) {
- nextI = end;
- }
- i = nextI;
}
+ i = nextTargetStart;
+
}
} catch (TaxonomySearchException e) {
throw new DrugSearchException("Problem with finidng information about organism", e);
@@ -382,6 +337,67 @@ public class DrugbankHTMLParser extends DrugAnnotation implements IExternalServi
return result;
}
+ /**
+ * Parse html info about target into {@link Target} structure.
+ *
+ * @param htmlPage
+ * string with html content
+ * @return {@link Target} for given html content
+ * @throws UniprotSearchException
+ * thrown when there is a problem with accessing uniprot db
+ * @throws TaxonomySearchException
+ * thrown when there is a problem with accessing taxonomy db
+ */
+ protected Target parseTarget(String htmlPage) throws UniprotSearchException, TaxonomySearchException {
+ int kindIndex = htmlPage.indexOf("<dt>Kind</dt><dd>");
+ kindIndex += "<dt>Kind</dt><dd>".length();
+ int endKindIndex = htmlPage.indexOf("</dd>");
+ String type = "";
+ if (kindIndex > 0 && endKindIndex > kindIndex) {
+ type = htmlPage.substring(kindIndex, endKindIndex);
+ }
+ if (type.trim().equalsIgnoreCase("Protein")) {
+ int uniprotIdStart = htmlPage.indexOf("/biodb/polypeptides/") + "/biodb/polypeptides/".length();
+ Target result = new Target();
+ result.setType(TargetType.SINGLE_PROTEIN);
+
+ // Getting ID && Name
+ int uniprotIdEnd = htmlPage.indexOf('"', uniprotIdStart);
+ String uniprotId = htmlPage.substring(uniprotIdStart, uniprotIdEnd);
+ MiriamData uniprotTarget = new MiriamData(MiriamType.UNIPROT, uniprotId);
+ MiriamData hgncTarget = uniprotAnnotator.uniProtToHgnc(uniprotTarget);
+ if (hgncTarget != null) {
+ result.addGene(hgncTarget);
+ } else {
+ result.addGene(uniprotTarget);
+ }
+
+ int nameStart = uniprotIdEnd + 2;
+ int nameEnd = htmlPage.indexOf("</", uniprotIdStart);
+ String name = StringEscapeUtils.unescapeHtml4(htmlPage.substring(nameStart, nameEnd));
+ result.setName(name);
+
+ // Getting Organism
+ int organismStart = htmlPage.indexOf("Organism</dt><dd>", nameEnd) + "Organism</dt><dd>".length();
+ int organismEnd = htmlPage.indexOf("</dd>", organismStart);
+ result.setOrganism(getTaxonomyBackend().getByName(htmlPage.substring(organismStart, organismEnd)));
+
+ // Getting References
+ int referencesStart = htmlPage.indexOf("<strong>References</strong>", organismEnd);
+ if (referencesStart > 0) {
+ int referencesEnd = Math.min(htmlPage.indexOf("Details</a></div>", referencesStart), htmlPage.length());
+ if (referencesEnd < 0) {
+ referencesEnd = htmlPage.length();
+ }
+ result.addReferences(getPubmedFromRef(htmlPage.substring(referencesStart, referencesEnd)));
+ }
+ return result;
+ } else {
+ logger.warn("Unknown target type: " + type + ". Skipping.");
+ return null;
+ }
+ }
+
/**
* Finds information about drug in drugbank database.
*
diff --git a/annotation/src/test/java/lcsb/mapviewer/annotation/services/DrugbankHTMLParserTest.java b/annotation/src/test/java/lcsb/mapviewer/annotation/services/DrugbankHTMLParserTest.java
index 5b0dafb3ff..3e53239be1 100644
--- a/annotation/src/test/java/lcsb/mapviewer/annotation/services/DrugbankHTMLParserTest.java
+++ b/annotation/src/test/java/lcsb/mapviewer/annotation/services/DrugbankHTMLParserTest.java
@@ -711,4 +711,13 @@ public class DrugbankHTMLParserTest extends AnnotationTestFunctions {
assertEquals(uniprotAnnotator, parser.getUniprotAnnotator());
}
+ @Test
+ public void parseTarget() throws Exception {
+ String content = super.readFile("testFiles/drugbank/target-html-part.html");
+ Target target = drugBankHTMLParser.parseTarget(content);
+ assertNotNull(target);
+ assertEquals(2, target.getReferences().size());
+ assertEquals("Galactoside O-acetyltransferase", target.getName());
+ }
+
}
diff --git a/annotation/testFiles/drugbank/target-html-part.html b/annotation/testFiles/drugbank/target-html-part.html
new file mode 100644
index 0000000000..904b85d584
--- /dev/null
+++ b/annotation/testFiles/drugbank/target-html-part.html
@@ -0,0 +1 @@
+</div><strong>1. <a href="/biodb/polypeptides/P07464">Galactoside O-acetyltransferase</a></strong></div><div class="panel-body"><dl class="dl-horizontal"><dt>Kind</dt><dd>Protein</dd><dt>Organism</dt><dd>Escherichia coli (strain K12)</dd><dt>Pharmacological action</dt><dd><strong class="label label-warning">unknown</strong></dd></dl><dl class="dl-horizontal"><dt>General Function:</dt><dd>Galactoside o-acetyltransferase activity</dd><dt>Specific Function:</dt><dd>May assist cellular detoxification by acetylating non-metabolizable pyranosides, thereby preventing their reentry into the cell.</dd><dt>Gene Name:</dt><dd>lacA</dd><dt>Uniprot ID:</dt><dd><a target="_blank" class="wishart-link-out" href="http://www.uniprot.org/uniprot/P07464">P07464 <span class="glyphicon glyphicon-new-window"> </span></a></dd><dt>Uniprot Name:</dt><dd>Galactoside O-acetyltransferase</dd><dt>Molecular Weight:</dt><dd>22798.89 Da</dd></dl><h5><strong>References</strong></h5><blockquote class="references"><ol class="cite-this-references"><li id="reference-A1713">Overington JP, Al-Lazikani B, Hopkins AL: How many drug targets are there? Nat Rev Drug Discov. 2006 Dec;5(12):993-6. [<a target="_blank" class="wishart-link-out" href="http://www.ncbi.nlm.nih.gov/pubmed/17139284">PubMed:17139284 <span class="glyphicon glyphicon-new-window"> </span></a>] </li><li id="reference-A1715">Imming P, Sinning C, Meyer A: Drugs, their targets and the nature and number of drug targets. Nat Rev Drug Discov. 2006 Oct;5(10):821-34. [<a target="_blank" class="wishart-link-out" href="http://www.ncbi.nlm.nih.gov/pubmed/17016423">PubMed:17016423 <span class="glyphicon glyphicon-new-window"> </span></a>] </li></ol></blockquote></div></div><div class="panel panel-default" id="BE0000574"><div class="panel-heading"><div class="pull-right"><a class="btn btn-primary btn-xs" href="/biodb/polypeptides/P04035"><span class="glyphicon glyphicon-list"> </span> Details</a>
\ No newline at end of file
--
GitLab