X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fext%2Fensembl%2FEnsemblLookup.java;h=877331df71d5fc063f8a9f4535a06137917cc082;hb=6fd29e7e2753c9066fa31b76ab71827f96d471ff;hp=6483401145320aa077daa1568664e0cb05db61a8;hpb=2e6299399bd798ad2959a2b80fd3ac2d2a1374df;p=jalview.git diff --git a/src/jalview/ext/ensembl/EnsemblLookup.java b/src/jalview/ext/ensembl/EnsemblLookup.java index 6483401..877331d 100644 --- a/src/jalview/ext/ensembl/EnsemblLookup.java +++ b/src/jalview/ext/ensembl/EnsemblLookup.java @@ -34,15 +34,13 @@ import org.json.simple.parser.JSONParser; import org.json.simple.parser.ParseException; /** - * A client for the Ensembl lookup REST endpoint; used to find the Parent gene - * identifier given a transcript identifier. + * A client for the Ensembl lookup REST endpoint, used to find the gene + * identifier given a gene, transcript or protein identifier. * * @author gmcarstairs - * */ public class EnsemblLookup extends EnsemblRestClient { - /** * Default constructor (to use rest.ensembl.org) */ @@ -77,17 +75,26 @@ public class EnsemblLookup extends EnsemblRestClient protected URL getUrl(List ids) throws MalformedURLException { String identifier = ids.get(0); - return getUrl(identifier); + return getUrl(identifier, null); } /** + * Gets the url for lookup of the given identifier, optionally with objectType + * also specified in the request + * * @param identifier + * @param objectType * @return */ - protected URL getUrl(String identifier) + protected URL getUrl(String identifier, String objectType) { String url = getDomain() + "/lookup/id/" + identifier - + "?content-type=application/json"; + + CONTENT_TYPE_JSON; + if (objectType != null) + { + url += "&" + OBJECT_TYPE + "=" + objectType; + } + try { return new URL(url); @@ -116,20 +123,33 @@ public class EnsemblLookup extends EnsemblRestClient } /** - * Calls the Ensembl lookup REST endpoint and retrieves the 'Parent' for the - * given identifier, or null if not found + * Returns the gene id related to the given identifier, which may be for a + * gene, transcript or protein * * @param identifier * @return */ - public String getParent(String identifier) + public String getGeneId(String identifier) + { + return getGeneId(identifier, null); + } + + /** + * Returns the gene id related to the given identifier (which may be for a + * gene, transcript or protein) + * + * @param identifier + * @param objectType + * @return + */ + public String getGeneId(String identifier, String objectType) { List ids = Arrays.asList(new String[] { identifier }); BufferedReader br = null; try { - URL url = getUrl(identifier); + URL url = getUrl(identifier, objectType); if (url != null) { br = getHttpResponse(url, ids); @@ -155,8 +175,10 @@ public class EnsemblLookup extends EnsemblRestClient } /** - * Parses "Parent" from the JSON response and returns the value, or null if - * not found + * Parses the JSON response and returns the gene identifier, or null if not + * found. If the returned object_type is Gene, returns the id, if Transcript + * returns the Parent. If it is Translation (peptide identifier), then the + * Parent is the transcript identifier, so we redo the search with this value. * * @param br * @return @@ -164,17 +186,33 @@ public class EnsemblLookup extends EnsemblRestClient */ protected String parseResponse(BufferedReader br) throws IOException { - String parent = null; + String geneId = null; JSONParser jp = new JSONParser(); try { JSONObject val = (JSONObject) jp.parse(br); - parent = val.get("Parent").toString(); + String type = val.get(OBJECT_TYPE).toString(); + if (OBJECT_TYPE_GENE.equalsIgnoreCase(type)) + { + // got the gene - just returns its id + geneId = val.get(ID).toString(); + } + else if (OBJECT_TYPE_TRANSCRIPT.equalsIgnoreCase(type)) + { + // got the transcript - return its (Gene) Parent + geneId = val.get(PARENT).toString(); + } + else if (OBJECT_TYPE_TRANSLATION.equalsIgnoreCase(type)) + { + // got the protein - get its Parent, restricted to type Transcript + String transcriptId = val.get(PARENT).toString(); + geneId = getGeneId(transcriptId, OBJECT_TYPE_TRANSCRIPT); + } } catch (ParseException e) { // ignore } - return parent; + return geneId; } }