/*
- * Jalview - A Sequence Alignment Editor and Viewer (Version 2.9)
- * Copyright (C) 2015 The Jalview Authors
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
*
* This file is part of Jalview.
*
*/
package jalview.ws.dbsources;
+import jalview.bin.Cache;
+import jalview.datamodel.Alignment;
import jalview.datamodel.AlignmentI;
import jalview.datamodel.DBRefEntry;
import jalview.datamodel.DBRefSource;
import jalview.datamodel.PDBEntry;
+import jalview.datamodel.Sequence;
import jalview.datamodel.SequenceFeature;
import jalview.datamodel.SequenceI;
-import jalview.datamodel.UniprotEntry;
-import jalview.datamodel.UniprotFile;
-import jalview.ws.ebi.EBIFetchClient;
-import jalview.ws.seqfetcher.DbSourceProxy;
+import jalview.schemes.ResidueProperties;
+import jalview.util.StringUtils;
import jalview.ws.seqfetcher.DbSourceProxyImpl;
-
-import java.io.File;
-import java.io.FileReader;
-import java.io.Reader;
+import jalview.xml.binding.uniprot.DbReferenceType;
+import jalview.xml.binding.uniprot.Entry;
+import jalview.xml.binding.uniprot.FeatureType;
+import jalview.xml.binding.uniprot.LocationType;
+import jalview.xml.binding.uniprot.PositionType;
+import jalview.xml.binding.uniprot.PropertyType;
+
+import java.io.InputStream;
+import java.net.URL;
+import java.net.URLConnection;
+import java.util.ArrayList;
+import java.util.List;
import java.util.Vector;
-import org.exolab.castor.xml.Unmarshaller;
+import javax.xml.bind.JAXBContext;
+import javax.xml.bind.JAXBException;
+import javax.xml.stream.FactoryConfigurationError;
+import javax.xml.stream.XMLInputFactory;
+import javax.xml.stream.XMLStreamException;
+import javax.xml.stream.XMLStreamReader;
import com.stevesoft.pat.Regex;
/**
+ * This class queries the Uniprot database for sequence data, unmarshals the
+ * returned XML, and converts it to Jalview Sequence records (including attached
+ * database references and sequence features)
+ *
* @author JimP
*
*/
-public class Uniprot extends DbSourceProxyImpl implements DbSourceProxy
+public class Uniprot extends DbSourceProxyImpl
{
+ private static final String DEFAULT_UNIPROT_DOMAIN = "https://www.uniprot.org";
private static final String BAR_DELIMITER = "|";
- private static final String NEWLINE = "\n";
-
- private static org.exolab.castor.mapping.Mapping map;
-
/**
* Constructor
*/
public Uniprot()
{
super();
- addDbSourceProperty(DBRefSource.SEQDB, DBRefSource.SEQDB);
- addDbSourceProperty(DBRefSource.PROTSEQDB);
- // addDbSourceProperty(DBRefSource.MULTIACC, new Integer(50));
+ }
+
+ private String getDomain()
+ {
+ return Cache.getDefault("UNIPROT_DOMAIN", DEFAULT_UNIPROT_DOMAIN);
}
/*
*
* @see jalview.ws.DbSourceProxy#getAccessionSeparator()
*/
+ @Override
public String getAccessionSeparator()
{
- return null; // ";";
+ return null;
}
/*
*
* @see jalview.ws.DbSourceProxy#getAccessionValidator()
*/
+ @Override
public Regex getAccessionValidator()
{
return new Regex("([A-Z]+[0-9]+[A-Z0-9]+|[A-Z0-9]+_[A-Z0-9]+)");
*
* @see jalview.ws.DbSourceProxy#getDbSource()
*/
+ @Override
public String getDbSource()
{
return DBRefSource.UNIPROT;
*
* @see jalview.ws.DbSourceProxy#getDbVersion()
*/
+ @Override
public String getDbVersion()
{
return "0"; // we really don't know what version we're on.
}
- /**
- * Reads a file containing the reply to the EBI Fetch Uniprot data query,
- * unmarshals it to a UniprotFile object, and returns the list of UniprotEntry
- * data models (mapped from <entry> elements)
- *
- * @param fileReader
- * @return
- */
- public Vector<UniprotEntry> getUniprotEntries(Reader fileReader)
- {
- UniprotFile uni = new UniprotFile();
- try
- {
- if (map == null)
- {
- // 1. Load the mapping information from the file
- map = new org.exolab.castor.mapping.Mapping(uni.getClass()
- .getClassLoader());
- java.net.URL url = getClass().getResource("/uniprot_mapping.xml");
- map.loadMapping(url);
- }
-
- // 2. Unmarshal the data
- Unmarshaller unmar = new Unmarshaller(uni);
- unmar.setIgnoreExtraElements(true);
- unmar.setMapping(map);
- if (fileReader != null)
- {
- uni = (UniprotFile) unmar.unmarshal(fileReader);
- }
- } catch (Exception e)
- {
- System.out.println("Error getUniprotEntries() " + e);
- }
-
- return uni.getUniprotEntries();
- }
-
/*
* (non-Javadoc)
*
* @see jalview.ws.DbSourceProxy#getSequenceRecords(java.lang.String[])
*/
+ @Override
public AlignmentI getSequenceRecords(String queries) throws Exception
{
startQuery();
queries = queries.toUpperCase().replaceAll(
"(UNIPROT\\|?|UNIPROT_|UNIREF\\d+_|UNIREF\\d+\\|?)", "");
AlignmentI al = null;
- EBIFetchClient ebi = new EBIFetchClient();
- // uniprotxml parameter required since december 2007
- // uniprotkb dbname changed introduced december 2008
- File file = ebi.fetchDataAsFile("uniprotkb:" + queries, "uniprotxml",
- null);
- Vector<UniprotEntry> entries = getUniprotEntries(new FileReader(file));
+ String downloadstring = getDomain() + "/uniprot/" + queries
+ + ".xml";
+
+ URL url = new URL(downloadstring);
+ URLConnection urlconn = url.openConnection();
+ InputStream istr = urlconn.getInputStream();
+ List<Entry> entries = getUniprotEntries(istr);
if (entries != null)
{
- /*
- * If Castor binding included sequence@length, we could guesstimate the
- * size of buffer to hold the alignment
- */
- StringBuffer result = new StringBuffer(128);
- // First, make the new sequences
- for (UniprotEntry entry : entries)
- {
- StringBuilder name = constructSequenceFastaHeader(entry);
-
- result.append(name).append(NEWLINE)
- .append(entry.getUniprotSequence().getContent())
- .append(NEWLINE);
- }
-
- // Then read in the features and apply them to the dataset
- al = parseResult(result.toString());
- if (al != null)
- {
- // Decorate the alignment with database entries.
- addUniprotXrefs(al, entries);
- }
- else
+ List<SequenceI> seqs = new ArrayList<>();
+ for (Entry entry : entries)
{
- results = result;
+ seqs.add(uniprotEntryToSequence(entry));
}
+ al = new Alignment(seqs.toArray(new SequenceI[seqs.size()]));
}
+
stopQuery();
return al;
} catch (Exception e)
{
- stopQuery();
throw (e);
+ } finally
+ {
+ stopQuery();
}
}
/**
- * Construct a Fasta-format sequence header by concatenating the source,
- * accession id(s) and name(s), delimited by '|', plus any protein names, now
- * with space rather than bar delimiter
+ * Converts an Entry object (bound from Uniprot XML) to a Jalview Sequence
*
* @param entry
* @return
*/
- public static StringBuilder constructSequenceFastaHeader(
- UniprotEntry entry)
+ SequenceI uniprotEntryToSequence(Entry entry)
{
- StringBuilder name = new StringBuilder(32);
- name.append(">UniProt/Swiss-Prot");
+ String id = getUniprotEntryId(entry);
+ String seqString = entry.getSequence().getValue();
+
+ /*
+ * for backwards compatibility with Castor processing,
+ * remove any internal spaces
+ */
+ if (seqString.indexOf(' ') > -1)
+ {
+ seqString = seqString.replace(" ", "");
+ }
+ SequenceI sequence = new Sequence(id,
+ seqString);
+ sequence.setDescription(getUniprotEntryDescription(entry));
+
+ /*
+ * add a 'self' DBRefEntry for each accession
+ */
+ final String dbVersion = getDbVersion();
+ List<DBRefEntry> dbRefs = new ArrayList<>();
for (String accessionId : entry.getAccession())
{
- name.append(BAR_DELIMITER);
- name.append(accessionId);
+ DBRefEntry dbRef = new DBRefEntry(DBRefSource.UNIPROT, dbVersion,
+ accessionId);
+ dbRefs.add(dbRef);
}
- for (String n : entry.getName())
+
+ /*
+ * add a DBRefEntry for each dbReference element in the XML;
+ * also add a PDBEntry if type="PDB";
+ * also add an EMBLCDS dbref if protein sequence id is given
+ * also add an Ensembl dbref " " " " " "
+ */
+ Vector<PDBEntry> pdbRefs = new Vector<>();
+ for (DbReferenceType dbref : entry.getDbReference())
{
- name.append(BAR_DELIMITER);
- name.append(n);
+ String type = dbref.getType();
+ DBRefEntry dbr = new DBRefEntry(type,
+ DBRefSource.UNIPROT + ":" + dbVersion, dbref.getId());
+ dbRefs.add(dbr);
+ if ("PDB".equals(type))
+ {
+ pdbRefs.add(new PDBEntry(dbr));
+ }
+ if ("EMBL".equals(type))
+ {
+ /*
+ * e.g. Uniprot accession Q9BXM7 has
+ * <dbReference type="EMBL" id="M19359">
+ * <property type="protein sequence ID" value="AAA40981.1"/>
+ * <property type="molecule type" value="Genomic_DNA"/>
+ * </dbReference>
+ */
+ String cdsId = getProperty(dbref.getProperty(),
+ "protein sequence ID");
+ if (cdsId != null && cdsId.trim().length() > 0)
+ {
+ // remove version
+ String[] vrs = cdsId.split("\\.");
+ String version = vrs.length > 1 ? vrs[1]
+ : DBRefSource.UNIPROT + ":" + dbVersion;
+ dbr = new DBRefEntry(DBRefSource.EMBLCDS, version, vrs[0]);
+ dbRefs.add(dbr);
+ }
+ }
+ if ("Ensembl".equals(type))
+ {
+ /*
+ * e.g. Uniprot accession Q9BXM7 has
+ * <dbReference type="Ensembl" id="ENST00000321556">
+ * <molecule id="Q9BXM7-1"/>
+ * <property type="protein sequence ID" value="ENSP00000364204"/>
+ * <property type="gene ID" value="ENSG00000158828"/>
+ * </dbReference>
+ */
+ String cdsId = getProperty(dbref.getProperty(),
+ "protein sequence ID");
+ if (cdsId != null && cdsId.trim().length() > 0)
+ {
+ dbr = new DBRefEntry(DBRefSource.ENSEMBL,
+ DBRefSource.UNIPROT + ":" + dbVersion, cdsId.trim());
+ dbRefs.add(dbr);
+ }
+ }
}
- if (entry.getProtein() != null && entry.getProtein().getName() != null)
+ /*
+ * create features; they have either begin and end, or position, in XML
+ */
+ sequence.setPDBId(pdbRefs);
+ if (entry.getFeature() != null)
{
- for (String nm : entry.getProtein().getName())
+ for (FeatureType uf : entry.getFeature())
{
- name.append(" ").append(nm);
+ LocationType location = uf.getLocation();
+ int start = 0;
+ int end = 0;
+ if (location.getPosition() != null)
+ {
+ start = location.getPosition().getPosition().intValue();
+ end = start;
+ }
+ else
+ {
+ start = location.getBegin().getPosition().intValue();
+ end = location.getEnd().getPosition().intValue();
+ }
+ SequenceFeature sf = new SequenceFeature(uf.getType(),
+ getDescription(uf), start, end, "Uniprot");
+ sf.setStatus(uf.getStatus());
+ sequence.addSequenceFeature(sf);
}
}
- return name;
+ for (DBRefEntry dbr : dbRefs)
+ {
+ sequence.addDBRef(dbr);
+ }
+ return sequence;
}
/**
- * add an ordered set of UniprotEntry objects to an ordered set of seuqences.
+ * A helper method that builds a sequence feature description
*
- * @param al
- * - a sequence of n sequences
- * @param entries
- * a list of n uniprot entries to be analysed.
+ * @param feature
+ * @return
*/
- public void addUniprotXrefs(AlignmentI al, Vector<UniprotEntry> entries)
+ static String getDescription(FeatureType feature)
{
- final String dbVersion = getDbVersion();
-
- for (int i = 0; i < entries.size(); i++)
+ String orig = feature.getOriginal();
+ List<String> variants = feature.getVariation();
+ StringBuilder sb = new StringBuilder();
+
+ /*
+ * append variant in standard format if present
+ * e.g. p.Arg59Lys
+ * multiple variants are split over lines using <br>
+ */
+ boolean asHtml = false;
+ if (orig != null && !orig.isEmpty() && variants != null
+ && !variants.isEmpty())
{
- UniprotEntry entry = entries.elementAt(i);
- Vector<PDBEntry> onlyPdbEntries = new Vector<PDBEntry>();
- Vector<DBRefEntry> dbxrefs = new Vector<DBRefEntry>();
-
- for (PDBEntry pdb : entry.getDbReference())
+ int p = 0;
+ for (String var : variants)
{
- DBRefEntry dbr = new DBRefEntry();
- dbr.setSource(pdb.getType());
- dbr.setAccessionId(pdb.getId());
- dbr.setVersion(DBRefSource.UNIPROT + ":" + dbVersion);
- dbxrefs.addElement(dbr);
- if ("PDB".equals(pdb.getType()))
+ // TODO proper HGVS nomenclature for delins structural variations
+ // http://varnomen.hgvs.org/recommendations/protein/variant/delins/
+ // for now we are pragmatic - any orig/variant sequence longer than
+ // three characters is shown with single-character notation rather than
+ // three-letter notation
+ sb.append("p.");
+ if (orig.length() < 4)
{
- onlyPdbEntries.addElement(pdb);
+ for (int c = 0, clen = orig.length(); c < clen; c++)
+ {
+ char origchar = orig.charAt(c);
+ String orig3 = ResidueProperties.aa2Triplet.get("" + origchar);
+ sb.append(orig3 == null ? origchar
+ : StringUtils.toSentenceCase(orig3));
+ }
+ }
+ else
+ {
+ sb.append(orig);
}
- }
- SequenceI sq = al.getSequenceAt(i);
- while (sq.getDatasetSequence() != null)
- {
- sq = sq.getDatasetSequence();
- }
+ LocationType location = feature.getLocation();
+ PositionType start = location.getPosition() == null
+ ? location.getBegin()
+ : location.getPosition();
+ sb.append(Integer.toString(start.getPosition().intValue()));
- for (String accessionId : entry.getAccession())
- {
- /*
- * add as uniprot whether retrieved from uniprot or uniprot_name
- */
- sq.addDBRef(new DBRefEntry(DBRefSource.UNIPROT, dbVersion,
- accessionId));
+ if (var.length() < 4)
+ {
+ for (int c = 0, clen = var.length(); c < clen; c++)
+ {
+ char varchar = var.charAt(c);
+ String var3 = ResidueProperties.aa2Triplet.get("" + varchar);
+
+ sb.append(var3 != null ? StringUtils.toSentenceCase(var3)
+ : "" + varchar);
+ }
+ }
+ else
+ {
+ sb.append(var);
+ }
+ if (++p != variants.size())
+ {
+ sb.append("<br/> ");
+ asHtml = true;
+ }
+ else
+ {
+ sb.append(" ");
+ }
}
+ }
+ String description = feature.getDescription();
+ if (description != null)
+ {
+ sb.append(description);
+ }
+ if (asHtml)
+ {
+ sb.insert(0, "<html>");
+ sb.append("</html>");
+ }
- for (DBRefEntry dbRef : dbxrefs)
- {
- sq.addDBRef(dbRef);
- }
- sq.setPDBId(onlyPdbEntries);
- if (entry.getFeature() != null)
+ return sb.toString();
+ }
+
+ /**
+ * A helper method that searches the list of properties for one with the given
+ * key, and if found returns the property value, else returns null
+ *
+ * @param properties
+ * @param key
+ * @return
+ */
+ static String getProperty(List<PropertyType> properties, String key)
+ {
+ String value = null;
+ if (properties != null)
+ {
+ for (PropertyType prop : properties)
{
- for (SequenceFeature sf : entry.getFeature())
+ if (key.equals(prop.getType()))
{
- sf.setFeatureGroup("Uniprot");
- sq.addSequenceFeature(sf);
+ value = prop.getValue();
+ break;
}
}
}
+ return value;
+ }
+
+ /**
+ * Extracts xml element entry/protein/recommendedName/fullName
+ *
+ * @param entry
+ * @return
+ */
+ static String getUniprotEntryDescription(Entry entry)
+ {
+ String desc = "";
+ if (entry.getProtein() != null
+ && entry.getProtein().getRecommendedName() != null)
+ {
+ // fullName is mandatory if recommendedName is present
+ desc = entry.getProtein().getRecommendedName().getFullName()
+ .getValue();
+ }
+ return desc;
+ }
+
+ /**
+ * Constructs a sequence id by concatenating all entry/name elements with '|'
+ * separator
+ *
+ * @param entry
+ * @return
+ */
+ static String getUniprotEntryId(Entry entry)
+ {
+ StringBuilder name = new StringBuilder(32);
+ for (String n : entry.getName())
+ {
+ if (name.length() > 0)
+ {
+ name.append(BAR_DELIMITER);
+ }
+ name.append(n);
+ }
+ return name.toString();
}
/*
*
* @see jalview.ws.DbSourceProxy#isValidReference(java.lang.String)
*/
+ @Override
public boolean isValidReference(String accession)
{
// TODO: make the following a standard validator
/**
* return LDHA_CHICK uniprot entry
*/
+ @Override
public String getTestQuery()
{
return "P00340";
}
+ @Override
public String getDbName()
{
return "Uniprot"; // getDbSource();
{
return 0;
}
+
+ /**
+ * Reads the reply to the EBI Fetch Uniprot data query, unmarshals it to an
+ * Uniprot object, and returns the enclosed Entry objects, or null on any
+ * failure
+ *
+ * @param is
+ * @return
+ */
+ public List<Entry> getUniprotEntries(InputStream is)
+ {
+ List<Entry> entries = null;
+ try
+ {
+ JAXBContext jc = JAXBContext
+ .newInstance("jalview.xml.binding.uniprot");
+ XMLStreamReader streamReader = XMLInputFactory.newInstance()
+ .createXMLStreamReader(is);
+ javax.xml.bind.Unmarshaller um = jc.createUnmarshaller();
+ jalview.xml.binding.uniprot.Uniprot uniprot = (jalview.xml.binding.uniprot.Uniprot) um.unmarshal(streamReader);
+ if (uniprot != null && !uniprot.getEntry().isEmpty())
+ {
+ entries = uniprot.getEntry();
+ }
+ } catch (JAXBException | XMLStreamException
+ | FactoryConfigurationError e)
+ {
+ e.printStackTrace();
+ }
+ return entries;
+ }
}
git://source.jalview.org / jalview.git / blobdiff