X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2Fgff%2FGff3Helper.java;h=56ec5ef8475df1312e596757b4e3860fc21bcaa6;hb=3b3c59cfa50e942d2fa5b367b7117cade9459ce7;hp=1ffe27cb8078f72bb4d93a55616b0efa94eb92de;hpb=f9b80711054b61e8c2257488a1637e15616cb9c9;p=jalview.git
diff --git a/src/jalview/io/gff/Gff3Helper.java b/src/jalview/io/gff/Gff3Helper.java
index 1ffe27cb..56ec5ef 100644
--- a/src/jalview/io/gff/Gff3Helper.java
+++ b/src/jalview/io/gff/Gff3Helper.java
@@ -1,3 +1,23 @@
+/*
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
+ *
+ * This file is part of Jalview.
+ *
+ * Jalview is free software: you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, either version 3
+ * of the License, or (at your option) any later version.
+ *
+ * Jalview is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ * PURPOSE. See the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with Jalview. If not, see .
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
+ */
package jalview.io.gff;
import jalview.datamodel.AlignedCodonFrame;
@@ -19,6 +39,8 @@ import java.util.Map;
*/
public class Gff3Helper extends GffHelperBase
{
+ public static final String ALLELES = "alleles";
+
protected static final String TARGET = "Target";
protected static final String ID = "ID";
@@ -69,15 +91,16 @@ public class Gff3Helper extends GffHelperBase
String atts = gff[ATTRIBUTES_COL];
Map> attributes = parseNameValuePairs(atts);
- if (SequenceOntology.getInstance().isProteinMatch(soTerm))
+ SequenceOntologyI so = SequenceOntologyFactory.getInstance();
+ if (so.isA(soTerm, SequenceOntologyI.PROTEIN_MATCH))
{
- sf = processProteinMatch(attributes, seq, gff, align,
- newseqs, relaxedIdMatching);
+ sf = processProteinMatch(attributes, seq, gff, align, newseqs,
+ relaxedIdMatching);
}
- else if (SequenceOntology.getInstance().isNucleotideMatch(soTerm))
+ else if (so.isA(soTerm, SequenceOntologyI.NUCLEOTIDE_MATCH))
{
- sf = processNucleotideMatch(attributes, seq, gff, align,
- newseqs, relaxedIdMatching);
+ sf = processNucleotideMatch(attributes, seq, gff, align, newseqs,
+ relaxedIdMatching);
}
else
{
@@ -91,7 +114,7 @@ public class Gff3Helper extends GffHelperBase
*/
sf = buildSequenceFeature(gff, null);
}
-
+
return sf;
}
@@ -118,8 +141,7 @@ public class Gff3Helper extends GffHelperBase
protected SequenceFeature processNucleotideMatch(
Map> attributes, SequenceI seq,
String[] gffColumns, AlignmentI align, List newseqs,
- boolean relaxedIdMatching)
- throws IOException
+ boolean relaxedIdMatching) throws IOException
{
String strand = gffColumns[STRAND_COL];
@@ -132,8 +154,8 @@ public class Gff3Helper extends GffHelperBase
*/
if ("-".equals(strand))
{
- System.err
- .println("Skipping mapping from reverse complement as not yet supported");
+ System.err.println(
+ "Skipping mapping from reverse complement as not yet supported");
return null;
}
@@ -165,8 +187,8 @@ public class Gff3Helper extends GffHelperBase
* (new or existing) virtual sequence in the newseqs list
*/
String targetId = findTargetId(tokens[0], attributes);
- SequenceI mappedSequence1 = findSequence(targetId, align,
- newseqs, relaxedIdMatching);
+ SequenceI mappedSequence1 = findSequence(targetId, align, newseqs,
+ relaxedIdMatching);
SequenceI mappedSequence = mappedSequence1;
if (mappedSequence == null)
{
@@ -194,8 +216,7 @@ public class Gff3Helper extends GffHelperBase
int fromStart = Integer.parseInt(gffColumns[START_COL]);
int fromEnd = Integer.parseInt(gffColumns[END_COL]);
MapList mapping = constructMappingFromAlign(fromStart, fromEnd,
- toStart, toEnd,
- MappingType.NucleotideToNucleotide);
+ toStart, toEnd, MappingType.NucleotideToNucleotide);
if (mapping != null)
{
@@ -225,7 +246,8 @@ public class Gff3Helper extends GffHelperBase
* @return
*/
@SuppressWarnings("unused")
- protected String findTargetId(String target, Map> set)
+ protected String findTargetId(String target,
+ Map> set)
{
return target;
}
@@ -256,8 +278,8 @@ public class Gff3Helper extends GffHelperBase
* @throws IOException
*/
protected SequenceFeature processProteinMatch(
- Map> set, SequenceI seq,
- String[] gffColumns, AlignmentI align, List newseqs,
+ Map> set, SequenceI seq, String[] gffColumns,
+ AlignmentI align, List newseqs,
boolean relaxedIdMatching)
{
// This is currently tailored to InterProScan GFF output:
@@ -279,8 +301,8 @@ public class Gff3Helper extends GffHelperBase
for (String target : targets)
{
- SequenceI mappedSequence1 = findSequence(findTargetId(target, set), align,
- newseqs, relaxedIdMatching);
+ SequenceI mappedSequence1 = findSequence(findTargetId(target, set),
+ align, newseqs, relaxedIdMatching);
SequenceI mappedSequence = mappedSequence1;
if (mappedSequence == null)
{
@@ -291,10 +313,9 @@ public class Gff3Helper extends GffHelperBase
* give the mapped sequence a copy of the sequence feature, with
* start/end range adjusted
*/
- SequenceFeature sf2 = new SequenceFeature(sf);
- sf2.setBegin(1);
int sequenceFeatureLength = 1 + sf.getEnd() - sf.getBegin();
- sf2.setEnd(sequenceFeatureLength);
+ SequenceFeature sf2 = new SequenceFeature(sf, 1,
+ sequenceFeatureLength, sf.getFeatureGroup(), sf.getScore());
mappedSequence.addSequenceFeature(sf2);
/*
@@ -302,8 +323,8 @@ public class Gff3Helper extends GffHelperBase
* renamed with its qualified accession id; renaming has to wait until
* all sequence reference resolution is complete
*/
- String accessionId = StringUtils.listToDelimitedString(
- set.get(NAME), ",");
+ String accessionId = StringUtils
+ .listToDelimitedString(set.get(NAME), ",");
if (accessionId.length() > 0)
{
String database = sf.getType(); // TODO InterProScan only??
@@ -343,9 +364,11 @@ public class Gff3Helper extends GffHelperBase
*/
@Override
protected SequenceFeature buildSequenceFeature(String[] gff,
+ int typeColumn, String group,
Map> attributes)
{
- SequenceFeature sf = super.buildSequenceFeature(gff, attributes);
+ SequenceFeature sf = super.buildSequenceFeature(gff, typeColumn, group,
+ attributes);
String desc = getDescription(sf, attributes);
if (desc != null)
{
@@ -371,23 +394,41 @@ public class Gff3Helper extends GffHelperBase
desc = target.split(" ")[0];
}
- if (SequenceOntology.getInstance().isSequenceVariant(sf.getType()))
+ SequenceOntologyI so = SequenceOntologyFactory.getInstance();
+ String type = sf.getType();
+ if (so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT))
{
/*
* Ensembl returns dna variants as 'alleles'
*/
- desc = StringUtils.listToDelimitedString(
- attributes.get("alleles"), ",");
+ desc = StringUtils.listToDelimitedString(attributes.get(ALLELES),
+ ",");
}
/*
- * Ensembl returns gene name as 'Name' for a transcript
+ * extract 'Name' for a transcript (to show gene name)
+ * or an exon (so 'colour by label' shows exon boundaries)
*/
- if (SequenceOntology.getInstance().isA(sf.getType(),
- SequenceOntology.TRANSCRIPT))
+ if (SequenceOntologyI.NMD_TRANSCRIPT_VARIANT.equals(type)
+ || so.isA(type, SequenceOntologyI.TRANSCRIPT)
+ || so.isA(type, SequenceOntologyI.EXON))
{
desc = StringUtils.listToDelimitedString(attributes.get("Name"), ",");
}
+
+ /*
+ * if the above fails, try ID
+ */
+ if (desc == null)
+ {
+ desc = (String) sf.getValue(ID);
+ }
+
+ /*
+ * and decode comma, equals, semi-colon as required by GFF3 spec
+ */
+ desc = StringUtils.urlDecode(desc, GFF_ENCODABLE);
+
return desc;
}
}