X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fext%2Fensembl%2FEnsemblLookup.java;h=c6b794aebdec283445a70c63127bcd663039fd7d;hb=f59dd9efbb3dfc313ab0b0507832e21cd0076fe1;hp=f31a3f09c29d3bca1920f221d387d7e95fe3973e;hpb=f87f3bc59e76f6112821e01ed88cf4c3ed853379;p=jalview.git diff --git a/src/jalview/ext/ensembl/EnsemblLookup.java b/src/jalview/ext/ensembl/EnsemblLookup.java index f31a3f0..c6b794a 100644 --- a/src/jalview/ext/ensembl/EnsemblLookup.java +++ b/src/jalview/ext/ensembl/EnsemblLookup.java @@ -20,13 +20,17 @@ */ package jalview.ext.ensembl; +import jalview.bin.Cache; import jalview.datamodel.AlignmentI; +import jalview.datamodel.GeneLociI; +import jalview.util.MapList; 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 org.json.simple.JSONObject; @@ -34,23 +38,15 @@ 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, or to extract the + * species or chromosomal coordinates from the same service response * * @author gmcarstairs - * */ public class EnsemblLookup extends EnsemblRestClient { - - private static final String OBJECT_TYPE_TRANSLATION = "Translation"; - private static final String PARENT = "Parent"; - private static final String OBJECT_TYPE_TRANSCRIPT = "Transcript"; - private static final String ID = "id"; - private static final String OBJECT_TYPE_GENE = "Gene"; - private static final String OBJECT_TYPE = "object_type"; - - private String lastId; + private static final String SPECIES = "species"; /** * Default constructor (to use rest.ensembl.org) @@ -86,17 +82,26 @@ 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_JSON; + if (objectType != null) + { + url += "&" + OBJECT_TYPE + "=" + objectType; + } + try { return new URL(url); @@ -112,48 +117,117 @@ public class EnsemblLookup extends EnsemblRestClient return true; } - @Override - protected String getRequestMimeType(boolean multipleIds) + /** + * Returns the gene id related to the given identifier (which may be for a + * gene, transcript or protein), or null if none is found + * + * @param identifier + * @return + */ + public String getGeneId(String identifier) { - return "application/json"; + return getGeneId(identifier, null); } - @Override - protected String getResponseMimeType() + /** + * Returns the gene id related to the given identifier (which may be for a + * gene, transcript or protein), or null if none is found + * + * @param identifier + * @param objectType + * @return + */ + public String getGeneId(String identifier, String objectType) { - return "application/json"; + return parseGeneId(getResult(identifier, objectType)); } /** - * Calls the Ensembl lookup REST endpoint and retrieves the 'Parent' for the + * 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 + */ + protected String parseGeneId(JSONObject val) + { + if (val == null) + { + return null; + } + String geneId = null; + String type = val.get(OBJECT_TYPE).toString(); + if (OBJECT_TYPE_GENE.equalsIgnoreCase(type)) + { + // got the gene - just returns its id + geneId = val.get(JSON_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); + } + + return geneId; + } + + /** + * Calls the Ensembl lookup REST endpoint and retrieves the 'species' for the * given identifier, or null if not found * * @param identifier * @return */ - public String getGeneId(String identifier) + public String getSpecies(String identifier) + { + String species = null; + JSONObject json = getResult(identifier, null); + if (json != null) + { + Object o = json.get(SPECIES); + if (o != null) + { + species = o.toString(); + } + } + return species; + } + + /** + * Calls the /lookup/id rest service and returns the response as a JSONObject, + * or null if any error + * + * @param identifier + * @param objectType + * (optional) + * @return + */ + protected JSONObject getResult(String identifier, String objectType) { List ids = Arrays.asList(new String[] { identifier }); BufferedReader br = null; try { - URL url = getUrl(identifier); - if (identifier.equals(lastId)) - { - System.err.println("** Ensembl lookup " + url.toString() - + " looping on Parent!"); - return null; - } - lastId = identifier; + URL url = getUrl(identifier, objectType); + if (url != null) { br = getHttpResponse(url, ids); } - return br == null ? null : parseResponse(br); - } catch (IOException e) + return br == null ? null : (JSONObject) (new JSONParser().parse(br)); + } catch (IOException | ParseException e) { - // ignore + System.err.println("Error parsing " + identifier + " lookup response " + + e.getMessage()); return null; } finally { @@ -171,53 +245,82 @@ public class EnsemblLookup extends EnsemblRestClient } /** - * 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. + * Calls the /lookup/id rest service for the given id, and if successful, + * parses and returns the gene's chromosomal coordinates * - * @param br + * @param geneId * @return - * @throws IOException */ - protected String parseResponse(BufferedReader br) throws IOException + public GeneLociI getGeneLoci(String geneId) { - String geneId = null; - JSONParser jp = new JSONParser(); + return parseGeneLoci(getResult(geneId, OBJECT_TYPE_GENE)); + } + + /** + * Parses the /lookup/id response for species, asssembly_name, + * seq_region_name, start, end and returns an object that wraps them, or null + * if unsuccessful + * + * @param json + * @return + */ + GeneLociI parseGeneLoci(JSONObject json) + { + if (json == null) + { + return null; + } + try { - JSONObject val = (JSONObject) jp.parse(br); - String type = val.get(OBJECT_TYPE).toString(); - if (OBJECT_TYPE_GENE.equalsIgnoreCase(type)) + final String species = json.get("species").toString(); + final String assembly = json.get("assembly_name").toString(); + final String chromosome = json.get("seq_region_name").toString(); + String strand = json.get("strand").toString(); + int start = Integer.parseInt(json.get("start").toString()); + int end = Integer.parseInt(json.get("end").toString()); + int fromEnd = end - start + 1; + boolean reverseStrand = "-1".equals(strand); + int toStart = reverseStrand ? end : start; + int toEnd = reverseStrand ? start : end; + List fromRange = Collections.singletonList(new int[] { 1, + fromEnd }); + List toRange = Collections.singletonList(new int[] { toStart, + toEnd }); + final MapList map = new MapList(fromRange, toRange, 1, 1); + return new GeneLociI() { - geneId = val.get(ID).toString(); - } - else if (OBJECT_TYPE_TRANSCRIPT.equalsIgnoreCase(type)) - { - geneId = val.get(PARENT).toString(); - } - else if (OBJECT_TYPE_TRANSLATION.equalsIgnoreCase(type)) - { - String transcriptId = val.get(PARENT).toString(); - try + + @Override + public String getSpeciesId() { - geneId = getGeneId(transcriptId); - } catch (StackOverflowError e) + return species == null ? "" : species; + } + + @Override + public String getAssemblyId() { - /* - * unlikely data condition error! - */ - System.err - .println("** Ensembl lookup " - + getUrl(transcriptId).toString() - + " looping on Parent!"); + return assembly; } - } - } catch (ParseException e) + + @Override + public String getChromosomeId() + { + return chromosome; + } + + @Override + public MapList getMap() + { + return map; + } + }; + } catch (NullPointerException | NumberFormatException e) { - // ignore + Cache.log.error("Error looking up gene loci: " + e.getMessage()); + e.printStackTrace(); } - return geneId; + return null; } }