import jalview.util.DBRefUtils;
import jalview.util.IntRangeComparator;
import jalview.util.MapList;
+import jalview.util.Platform;
-import java.io.BufferedReader;
import java.io.IOException;
import java.net.MalformedURLException;
import java.net.URL;
import java.util.Arrays;
import java.util.Collections;
import java.util.List;
+import java.util.Map;
-import org.json.simple.JSONObject;
-import org.json.simple.parser.JSONParser;
import org.json.simple.parser.ParseException;
/**
*/
public abstract class EnsemblSeqProxy extends EnsemblRestClient
{
- protected static final String NAME = "Name";
-
protected static final String DESCRIPTION = "description";
/*
String accId = querySeq.getName();
try
{
+ System.out.println("Adding protein product for " + accId);
AlignmentI protein = new EnsemblProtein(getDomain())
.getSequenceRecords(accId);
if (protein == null || protein.getHeight() == 0)
inProgress = false;
throw new JalviewException("ENSEMBL Rest API not available.");
}
- BufferedReader br = getSequenceReader(ids);
- if (br == null)
- {
- return alignment;
- }
+ Platform.timeCheck("EnsemblSeqProx.fetchSeq ", Platform.TIME_MARK);
- List<SequenceI> seqs = parseSequenceJson(br);
+ List<SequenceI> seqs = parseSequenceJson(ids);
+ if (seqs == null)
+ return alignment;
if (seqs.isEmpty())
{
}
/**
- * Parses a JSON response into a list of sequences
+ * Parses a JSON response for a single sequence ID query
*
* @param br
- * @return
+ * @return a single jalview.datamodel.Sequence
* @see http://rest.ensembl.org/documentation/info/sequence_id
*/
- protected List<SequenceI> parseSequenceJson(BufferedReader br)
+ @SuppressWarnings("unchecked")
+ protected List<SequenceI> parseSequenceJson(List<String> ids)
{
- JSONParser jp = new JSONParser();
List<SequenceI> result = new ArrayList<>();
try
{
* for now, assumes only one sequence returned; refactor if needed
* in future to handle a JSONArray with more than one
*/
- final JSONObject val = (JSONObject) jp.parse(br);
+ Map<String, Object> val = (Map<String, Object>) getJSON(null, ids, -1, MODE_MAP, null);
+ if (val == null)
+ return null;
Object s = val.get("desc");
String desc = s == null ? null : s.toString();
s = val.get("id");
// @see https://github.com/Ensembl/ensembl-rest/wiki/Output-formats
urlstring.append("?type=").append(getSourceEnsemblType().getType());
urlstring.append(("&Accept=application/json"));
- urlstring.append(("&Content-Type=application/json"));
+ urlstring.append(("&content-type=application/json"));
String objectType = getObjectType();
if (objectType != null)
protected MapList getGenomicRangesFromFeatures(SequenceI sourceSequence,
String accId, int start)
{
- List<SequenceFeature> sfs = sourceSequence.getFeatures()
- .getPositionalFeatures();
+ List<SequenceFeature> sfs = getIdentifyingFeatures(sourceSequence,
+ accId);
if (sfs.isEmpty())
{
return null;
for (SequenceFeature sf : sfs)
{
+ int strand = sf.getStrand();
+ strand = strand == 0 ? 1 : strand; // treat unknown as forward
+
+ if (directionSet && strand != direction)
+ {
+ // abort - mix of forward and backward
+ System.err
+ .println("Error: forward and backward strand for " + accId);
+ return null;
+ }
+ direction = strand;
+ directionSet = true;
+
/*
- * accept the target feature type or a specialisation of it
- * (e.g. coding_exon for exon)
+ * add to CDS ranges, semi-sorted forwards/backwards
*/
- if (identifiesSequence(sf, accId))
+ if (strand < 0)
{
- int strand = sf.getStrand();
- strand = strand == 0 ? 1 : strand; // treat unknown as forward
-
- if (directionSet && strand != direction)
- {
- // abort - mix of forward and backward
- System.err.println(
- "Error: forward and backward strand for " + accId);
- return null;
- }
- direction = strand;
- directionSet = true;
-
- /*
- * add to CDS ranges, semi-sorted forwards/backwards
- */
- if (strand < 0)
- {
- regions.add(0, new int[] { sf.getEnd(), sf.getBegin() });
- }
- else
- {
- regions.add(new int[] { sf.getBegin(), sf.getEnd() });
- }
- mappedLength += Math.abs(sf.getEnd() - sf.getBegin() + 1);
-
- if (!isSpliceable())
- {
- /*
- * 'gene' sequence is contiguous so we can stop as soon as its
- * identifying feature has been found
- */
- break;
- }
+ regions.add(0, new int[] { sf.getEnd(), sf.getBegin() });
+ }
+ else
+ {
+ regions.add(new int[] { sf.getBegin(), sf.getEnd() });
}
+ mappedLength += Math.abs(sf.getEnd() - sf.getBegin() + 1);
}
if (regions.isEmpty())
}
/**
- * Answers true if the sequence being retrieved may occupy discontiguous
- * regions on the genomic sequence.
- */
- protected boolean isSpliceable()
- {
- return true;
- }
-
- /**
- * Returns true if the sequence feature marks positions of the genomic
- * sequence feature which are within the sequence being retrieved. For
- * example, an 'exon' feature whose parent is the target transcript marks the
- * cdna positions of the transcript.
- *
- * @param sf
- * @param accId
- * @return
- */
- protected abstract boolean identifiesSequence(SequenceFeature sf,
- String accId);
-
- /**
* Answers a list of sequence features that mark positions of the genomic
* sequence feature which are within the sequence being retrieved. For
* example, an 'exon' feature whose parent is the target transcript marks the
protected boolean featureMayBelong(SequenceFeature sf, String identifier)
{
String parent = (String) sf.getValue(PARENT);
- // using contains to allow for prefix "gene:", "transcript:" etc
if (parent != null
- && !parent.toUpperCase().contains(identifier.toUpperCase()))
+ && !parent.equalsIgnoreCase(identifier))
{
// this genomic feature belongs to a different transcript
return false;
return true;
}
+ /**
+ * Answers a short description of the sequence fetcher
+ */
@Override
public String getDescription()
{
/**
* Answers true if the feature type is either 'NMD_transcript_variant' or
- * 'transcript' or one of its sub-types in the Sequence Ontology. This is
- * needed because NMD_transcript_variant behaves like 'transcript' in Ensembl
+ * 'transcript' (or one of its sub-types in the Sequence Ontology). This is
+ * because NMD_transcript_variant behaves like 'transcript' in Ensembl
* although strictly speaking it is not (it is a sub-type of
* sequence_variant).
+ * <p>
+ * (This test was needed when fetching transcript features as GFF. As we are
+ * now fetching as JSON, all features have type 'transcript' so the check for
+ * NMD_transcript_variant is redundant. Left in for any future case arising.)
*
* @param featureType
* @return