*/
package jalview.util;
--import jalview.datamodel.DBRefEntry;
--import jalview.datamodel.DBRefSource;
- import jalview.datamodel.PDBEntry;
- import jalview.datamodel.SequenceI;
-
-import jalview.datamodel.Mapping;
-import jalview.datamodel.PDBEntry;
-import jalview.datamodel.SequenceI;
-
import java.util.ArrayList;
- import java.util.Arrays;
+ import java.util.BitSet;
import java.util.HashMap;
import java.util.HashSet;
import java.util.List;
import java.util.Map;
import com.stevesoft.pat.Regex;
++import jalview.datamodel.DBRefEntry;
++import jalview.datamodel.DBRefSource;
++import jalview.datamodel.Mapping;
++import jalview.datamodel.PDBEntry;
++import jalview.datamodel.SequenceI;
++
/**
* Utilities for handling DBRef objects and their collections.
*/
return rfs;
}
- interface DbRefComp
- {
- public boolean matches(DBRefEntry refa, DBRefEntry refb);
- }
-
- /**
- * match on all non-null fields in refa
- */
- // TODO unused - remove?
- public static DbRefComp matchNonNullonA = new DbRefComp()
- {
- @Override
- public boolean matches(DBRefEntry refa, DBRefEntry refb)
- {
- if (refa.getSource() == null
- || DBRefUtils.getCanonicalName(refb.getSource()).equals(
- DBRefUtils.getCanonicalName(refa.getSource())))
- {
- if (refa.getVersion() == null
- || refb.getVersion().equals(refa.getVersion()))
- {
- if (refa.getAccessionId() == null
- || refb.getAccessionId().equals(refa.getAccessionId()))
- {
- if (refa.getMap() == null || (refb.getMap() != null
- && refb.getMap().equals(refa.getMap())))
- {
- return true;
- }
- }
- }
- }
- return false;
- }
- };
-
- /**
- * either field is null or field matches for all of source, version, accession
- * id and map.
- */
- // TODO unused - remove?
- public static DbRefComp matchEitherNonNull = new DbRefComp()
- {
- @Override
- public boolean matches(DBRefEntry refa, DBRefEntry refb)
- {
- if (nullOrEqualSource(refa.getSource(), refb.getSource())
- && nullOrEqual(refa.getVersion(), refb.getVersion())
- && nullOrEqual(refa.getAccessionId(), refb.getAccessionId())
- && nullOrEqual(refa.getMap(), refb.getMap()))
- {
- return true;
- }
- return false;
- }
- };
-
- /**
- * accession ID and DB must be identical. Version is ignored. Map is either
- * not defined or is a match (or is compatible?)
- */
- // TODO unused - remove?
- public static DbRefComp matchDbAndIdAndEitherMap = new DbRefComp()
- {
- @Override
- public boolean matches(DBRefEntry refa, DBRefEntry refb)
- {
- if (refa.getSource() != null && refb.getSource() != null
- && DBRefUtils.getCanonicalName(refb.getSource()).equals(
- DBRefUtils.getCanonicalName(refa.getSource())))
- {
- // We dont care about version
- if (refa.getAccessionId() != null && refb.getAccessionId() != null
- // FIXME should be && not || here?
- || refb.getAccessionId().equals(refa.getAccessionId()))
- {
- if ((refa.getMap() == null || refb.getMap() == null)
- || (refa.getMap() != null && refb.getMap() != null
- && refb.getMap().equals(refa.getMap())))
- {
- return true;
- }
- }
- }
- return false;
- }
- };
-
- /**
- * accession ID and DB must be identical. Version is ignored. No map on either
- * or map but no maplist on either or maplist of map on a is the complement of
- * maplist of map on b.
- */
- // TODO unused - remove?
- public static DbRefComp matchDbAndIdAndComplementaryMapList = new DbRefComp()
- {
- @Override
- public boolean matches(DBRefEntry refa, DBRefEntry refb)
- {
- if (refa.getSource() != null && refb.getSource() != null
- && DBRefUtils.getCanonicalName(refb.getSource()).equals(
- DBRefUtils.getCanonicalName(refa.getSource())))
- {
- // We dont care about version
- if (refa.getAccessionId() != null && refb.getAccessionId() != null
- || refb.getAccessionId().equals(refa.getAccessionId()))
- {
- if ((refa.getMap() == null && refb.getMap() == null)
- || (refa.getMap() != null && refb.getMap() != null))
- {
- if ((refb.getMap().getMap() == null
- && refa.getMap().getMap() == null)
- || (refb.getMap().getMap() != null
- && refa.getMap().getMap() != null
- && refb.getMap().getMap().getInverse()
- .equals(refa.getMap().getMap())))
- {
- return true;
- }
- }
- }
- }
- return false;
- }
- };
-
- /**
- * accession ID and DB must be identical. Version is ignored. No map on both
- * or or map but no maplist on either or maplist of map on a is equivalent to
- * the maplist of map on b.
- */
- // TODO unused - remove?
- public static DbRefComp matchDbAndIdAndEquivalentMapList = new DbRefComp()
- {
- @Override
- public boolean matches(DBRefEntry refa, DBRefEntry refb)
- {
- if (refa.getSource() != null && refb.getSource() != null
- && DBRefUtils.getCanonicalName(refb.getSource()).equals(
- DBRefUtils.getCanonicalName(refa.getSource())))
- {
- // We dont care about version
- // if ((refa.getVersion()==null || refb.getVersion()==null)
- // || refb.getVersion().equals(refa.getVersion()))
- // {
- if (refa.getAccessionId() != null && refb.getAccessionId() != null
- || refb.getAccessionId().equals(refa.getAccessionId()))
- {
- if (refa.getMap() == null && refb.getMap() == null)
- {
- return true;
- }
- if (refa.getMap() != null && refb.getMap() != null
- && ((refb.getMap().getMap() == null
- && refa.getMap().getMap() == null)
- || (refb.getMap().getMap() != null
- && refa.getMap().getMap() != null
- && refb.getMap().getMap()
- .equals(refa.getMap().getMap()))))
- {
- return true;
- }
- }
- }
- return false;
- }
- };
-
- /**
- * accession ID and DB must be identical, or null on a. Version is ignored. No
- * map on either or map but no maplist on either or maplist of map on a is
- * equivalent to the maplist of map on b.
- */
- public static DbRefComp matchDbAndIdAndEitherMapOrEquivalentMapList = new DbRefComp()
- {
- @Override
- public boolean matches(DBRefEntry refa, DBRefEntry refb)
- {
- if (refa.getSource() != null && refb.getSource() != null
- && DBRefUtils.getCanonicalName(refb.getSource()).equals(
- DBRefUtils.getCanonicalName(refa.getSource())))
- {
- // We dont care about version
-
- if (refa.getAccessionId() == null
- || refa.getAccessionId().equals(refb.getAccessionId()))
- {
- if (refa.getMap() == null || refb.getMap() == null)
- {
- return true;
- }
- if ((refa.getMap() != null && refb.getMap() != null)
- && (refb.getMap().getMap() == null
- && refa.getMap().getMap() == null)
- || (refb.getMap().getMap() != null
- && refa.getMap().getMap() != null
- && (refb.getMap().getMap()
- .equals(refa.getMap().getMap()))))
- {
- return true;
- }
- }
- }
- return false;
- }
- };
-
- /**
- * accession ID only must be identical.
- */
- public static DbRefComp matchId = new DbRefComp()
- {
- @Override
- public boolean matches(DBRefEntry refa, DBRefEntry refb)
- {
- if (refa.getAccessionId() != null && refb.getAccessionId() != null
- && refb.getAccessionId().equals(refa.getAccessionId()))
- {
- return true;
- }
- return false;
- }
- };
+ /**
+ * look up source in an internal list of database reference sources and return
+ * the canonical jalview name for the source, or the original string if it has
+ * no canonical form.
+ *
+ * @param source
+ * @return canonical jalview source (one of jalview.datamodel.DBRefSource.*) or
+ * original source
+ */
+ public static String getCanonicalName(String source)
+ {
+ if (source == null)
+ {
+ return null;
+ }
+ String canonical = canonicalSourceNameLookup.get(source.toLowerCase());
+ return canonical == null ? source : canonical;
+ }
+
+ /**
+ * Returns a (possibly empty) list of those references that match the given
+ * entry. Currently uses a comparator which matches if
+ * <ul>
+ * <li>database sources are the same</li>
+ * <li>accession ids are the same</li>
+ * <li>both have no mapping, or the mappings are the same</li>
+ * </ul>
+ *
+ * @param ref Set of references to search
+ * @param entry pattern to match
+ * @param mode SEARCH_MODE_FULL for all; SEARCH_MODE_NO_MAP_NO_VERSION optional
+ * @return
+ */
+ public static List<DBRefEntry> searchRefs(List<DBRefEntry> ref, DBRefEntry entry, int mode) {
+ return searchRefs(ref, entry, matchDbAndIdAndEitherMapOrEquivalentMapList, mode);
+ }
+
+ /**
+ * Returns a list of those references that match the given accession id
+ * <ul>
+ * <li>database sources are the same</li>
+ * <li>accession ids are the same</li>
+ * <li>both have no mapping, or the mappings are the same</li>
+ * </ul>
+ *
+ * @param refs Set of references to search
+ * @param accId accession id to match
+ * @return
+ */
+ public static List<DBRefEntry> searchRefs(List<DBRefEntry> refs, String accId) {
+ List<DBRefEntry> rfs = new ArrayList<DBRefEntry>();
+ if (refs == null || accId == null) {
+ return rfs;
+ }
+ for (int i = 0, n = refs.size(); i < n; i++) {
+ DBRefEntry e = refs.get(i);
+ if (accId.equals(e.getAccessionId())) {
+ rfs.add(e);
+ }
+ }
+ return rfs;
+ // return searchRefs(refs, new DBRefEntry("", "", accId), matchId, SEARCH_MODE_FULL);
+ }
+
+ /**
+ * Returns a (possibly empty) list of those references that match the given
+ * entry, according to the given comparator.
+ *
+ * @param refs an array of database references to search
+ * @param entry an entry to compare against
+ * @param comparator
+ * @param mode SEARCH_MODE_FULL for all; SEARCH_MODE_NO_MAP_NO_VERSION
+ * optional
+ * @return
+ */
+ static List<DBRefEntry> searchRefs(List<DBRefEntry> refs, DBRefEntry entry, DbRefComp comparator, int mode) {
+ List<DBRefEntry> rfs = new ArrayList<DBRefEntry>();
+ if (refs == null || entry == null) {
+ return rfs;
+ }
+ for (int i = 0, n = refs.size(); i < n; i++) {
+ DBRefEntry e = refs.get(i);
+ if (comparator.matches(entry, e, SEARCH_MODE_FULL)) {
+ rfs.add(e);
+ }
+ }
+ return rfs;
+ }
+
+ interface DbRefComp {
+ default public boolean matches(DBRefEntry refa, DBRefEntry refb) {
+ return matches(refa, refb, SEARCH_MODE_FULL);
+ };
+
+ public boolean matches(DBRefEntry refa, DBRefEntry refb, int mode);
+ }
+
+ /**
+ * match on all non-null fields in refa
+ */
+ // TODO unused - remove? would be broken by equating "" with null
+ public static DbRefComp matchNonNullonA = new DbRefComp() {
+ @Override
+ public boolean matches(DBRefEntry refa, DBRefEntry refb, int mode) {
+ if ((mode & DB_SOURCE) != 0 &&
+ (refa.getSource() == null || DBRefUtils.getCanonicalName(refb.getSource())
+ .equals(DBRefUtils.getCanonicalName(refa.getSource())))) {
+ if ((mode & DB_VERSION) != 0 &&
+ (refa.getVersion() == null || refb.getVersion().equals(refa.getVersion()))) {
+ if ((mode & DB_ID) != 0 &&
+ (refa.getAccessionId() == null || refb.getAccessionId().equals(refa.getAccessionId()))) {
+ if ((mode & DB_MAP) != 0 &&
+ (refa.getMap() == null || (refb.getMap() != null && refb.getMap().equals(refa.getMap())))) {
+ return true;
+ }
+ }
+ }
+ }
+ return false;
+ }
+ };
+
+ /**
+ * either field is null or field matches for all of source, version, accession
+ * id and map.
+ */
+ // TODO unused - remove?
+ public static DbRefComp matchEitherNonNull = new DbRefComp() {
+ @Override
+ public boolean matches(DBRefEntry refa, DBRefEntry refb, int mode) {
+ if (nullOrEqualSource(refa.getSource(), refb.getSource())
+ && nullOrEqual(refa.getVersion(), refb.getVersion())
+ && nullOrEqual(refa.getAccessionId(), refb.getAccessionId())
+ && nullOrEqual(refa.getMap(), refb.getMap())) {
+ return true;
+ }
+ return false;
+ }
+
+ };
+ /**
+ * Parses a DBRefEntry and adds it to the sequence, also a PDBEntry if the
+ * database is PDB.
+ * <p>
+ * Used by file parsers to generate DBRefs from annotation within file (eg
+ * Stockholm)
+ *
+ * @param dbname
+ * @param version
+ * @param acn
+ * @param seq
+ * where to annotate with reference
+ * @return parsed version of entry that was added to seq (if any)
+ */
+ public static DBRefEntry parseToDbRef(SequenceI seq, String dbname,
+ String version, String acn)
+ {
+ DBRefEntry ref = null;
+ if (dbname != null)
+ {
+ String locsrc = DBRefUtils.getCanonicalName(dbname);
+ if (locsrc.equals(DBRefSource.PDB))
+ {
+ /*
+ * Check for PFAM style stockhom PDB accession id citation e.g.
+ * "1WRI A; 7-80;"
+ */
+ Regex r = new com.stevesoft.pat.Regex(
+ "([0-9][0-9A-Za-z]{3})\\s*(.?)\\s*;\\s*([0-9]+)-([0-9]+)");
+ if (r.search(acn.trim()))
+ {
+ String pdbid = r.stringMatched(1);
+ String chaincode = r.stringMatched(2);
+ if (chaincode == null)
+ {
+ chaincode = " ";
+ }
+ // String mapstart = r.stringMatched(3);
+ // String mapend = r.stringMatched(4);
+ if (chaincode.equals(" "))
+ {
+ chaincode = "_";
+ }
+ // construct pdb ref.
+ ref = new DBRefEntry(locsrc, version, pdbid + chaincode);
+ PDBEntry pdbr = new PDBEntry();
+ pdbr.setId(pdbid);
+ pdbr.setType(PDBEntry.Type.PDB);
+ pdbr.setChainCode(chaincode);
+ seq.addPDBId(pdbr);
+ }
+ else
+ {
+ System.err.println("Malformed PDB DR line:" + acn);
+ }
+ }
+ else
+ {
+ // default:
+ ref = new DBRefEntry(locsrc, version, acn.trim());
+ }
+ }
+ if (ref != null)
+ {
+ seq.addDBRef(ref);
+ }
+ return ref;
+ }
+
- /**
- * Returns true if either object is null, or they are equal
- *
- * @param o1
- * @param o2
- * @return
- */
- public static boolean nullOrEqual(Object o1, Object o2)
- {
- if (o1 == null || o2 == null)
- {
- return true;
- }
- return o1.equals(o2);
- }
-
- /**
- * canonicalise source string before comparing. null is always wildcard
- *
- * @param o1
- * - null or source string to compare
- * @param o2
- * - null or source string to compare
- * @return true if either o1 or o2 are null, or o1 equals o2 under
- * DBRefUtils.getCanonicalName
- * (o1).equals(DBRefUtils.getCanonicalName(o2))
- */
- public static boolean nullOrEqualSource(String o1, String o2)
- {
- if (o1 == null || o2 == null)
- {
- return true;
- }
- return DBRefUtils.getCanonicalName(o1)
- .equals(DBRefUtils.getCanonicalName(o2));
- }
-
- /**
- * Selects just the DNA or protein references from a set of references
- *
- * @param selectDna
- * if true, select references to 'standard' DNA databases, else to
- * 'standard' peptide databases
- * @param refs
- * a set of references to select from
- * @return
- */
- public static DBRefEntry[] selectDbRefs(boolean selectDna,
- DBRefEntry[] refs)
- {
- return selectRefs(refs,
- selectDna ? DBRefSource.DNACODINGDBS : DBRefSource.PROTEINDBS);
- // could attempt to find other cross
- // refs here - ie PDB xrefs
- // (not dna, not protein seq)
- }
-
- /**
+ /**
+ * accession ID and DB must be identical. Version is ignored. Map is either not
+ * defined or is a match (or is compatible?)
+ */
+ // TODO unused - remove?
+ public static DbRefComp matchDbAndIdAndEitherMap = new DbRefComp() {
+ @Override
+ public boolean matches(DBRefEntry refa, DBRefEntry refb, int mode) {
+ if (refa.getSource() != null && refb.getSource() != null && DBRefUtils.getCanonicalName(refb.getSource())
+ .equals(DBRefUtils.getCanonicalName(refa.getSource()))) {
+ // We dont care about version
+ if (refa.getAccessionId() != null && refb.getAccessionId() != null
+ // FIXME should be && not || here?
+ || refb.getAccessionId().equals(refa.getAccessionId())) {
+ if ((refa.getMap() == null || refb.getMap() == null) || (refa.getMap() != null
+ && refb.getMap() != null && refb.getMap().equals(refa.getMap()))) {
+ return true;
+ }
+ }
+ }
+ return false;
+ }
+ };
+
+ /**
+ * accession ID and DB must be identical. Version is ignored. No map on either
+ * or map but no maplist on either or maplist of map on a is the complement of
+ * maplist of map on b.
+ */
+ // TODO unused - remove?
+ public static DbRefComp matchDbAndIdAndComplementaryMapList = new DbRefComp() {
+ @Override
+ public boolean matches(DBRefEntry refa, DBRefEntry refb, int mode) {
+ if (refa.getSource() != null && refb.getSource() != null && DBRefUtils.getCanonicalName(refb.getSource())
+ .equals(DBRefUtils.getCanonicalName(refa.getSource()))) {
+ // We dont care about version
+ if (refa.getAccessionId() != null && refb.getAccessionId() != null
+ || refb.getAccessionId().equals(refa.getAccessionId())) {
+ if ((refa.getMap() == null && refb.getMap() == null)
+ || (refa.getMap() != null && refb.getMap() != null)) {
+ if ((refb.getMap().getMap() == null && refa.getMap().getMap() == null)
+ || (refb.getMap().getMap() != null && refa.getMap().getMap() != null
+ && refb.getMap().getMap().getInverse().equals(refa.getMap().getMap()))) {
+ return true;
+ }
+ }
+ }
+ }
+ return false;
+ }
+ };
+
+ /**
+ * accession ID and DB must be identical. Version is ignored. No map on both or
+ * or map but no maplist on either or maplist of map on a is equivalent to the
+ * maplist of map on b.
+ */
+ // TODO unused - remove?
+ public static DbRefComp matchDbAndIdAndEquivalentMapList = new DbRefComp() {
+ @Override
+ public boolean matches(DBRefEntry refa, DBRefEntry refb, int mode) {
+ if (refa.getSource() != null && refb.getSource() != null && DBRefUtils.getCanonicalName(refb.getSource())
+ .equals(DBRefUtils.getCanonicalName(refa.getSource()))) {
+ // We dont care about version
+ // if ((refa.getVersion()==null || refb.getVersion()==null)
+ // || refb.getVersion().equals(refa.getVersion()))
+ // {
+ if (refa.getAccessionId() != null && refb.getAccessionId() != null
+ || refb.getAccessionId().equals(refa.getAccessionId())) {
+ if (refa.getMap() == null && refb.getMap() == null) {
+ return true;
+ }
+ if (refa.getMap() != null && refb.getMap() != null
+ && ((refb.getMap().getMap() == null && refa.getMap().getMap() == null)
+ || (refb.getMap().getMap() != null && refa.getMap().getMap() != null
+ && refb.getMap().getMap().equals(refa.getMap().getMap())))) {
+ return true;
+ }
+ }
+ }
+ return false;
+ }
+ };
+
+ /**
+ * accession ID and DB must be identical, or null on a. Version is ignored. No
+ * map on either or map but no maplist on either or maplist of map on a is
+ * equivalent to the maplist of map on b.
+ */
+ public static DbRefComp matchDbAndIdAndEitherMapOrEquivalentMapList = new DbRefComp()
+ {
+ @Override
+ public boolean matches(DBRefEntry refa, DBRefEntry refb, int mode)
+ {
+ if (refa.getSource() != null && refb.getSource() != null && DBRefUtils.getCanonicalName(refb.getSource())
+ .equals(DBRefUtils.getCanonicalName(refa.getSource())))
+ {
+ // We dont care about version
+ if (refa.getAccessionId() == null || refa.getAccessionId().equals(refb.getAccessionId()))
+ {
+ if (refa.getMap() == null || refb.getMap() == null)
+ {
+ return true;
+ }
+ if ((refa.getMap() != null && refb.getMap() != null)
+ && (refb.getMap().getMap() == null && refa.getMap().getMap() == null)
+ || (refb.getMap().getMap() != null && refa.getMap().getMap() != null
+ && (refb.getMap().getMap().equals(refa.getMap().getMap()))))
+ {
+ return true;
+ }
+ }
+ }
+ return false;
+ }
+ };
+
+ /**
* Returns the (possibly empty) list of those supplied dbrefs which have the
* specified source database, with a case-insensitive match of source name
*
return matches;
}
- /**
- * promote direct database references to primary for nucleotide or protein
- * sequences if they have an appropriate primary ref
- * <table>
- * <tr>
- * <th>Seq Type</th>
- * <th>Primary DB</th>
- * <th>Direct which will be promoted</th>
- * </tr>
- * <tr align=center>
- * <td>peptides</td>
- * <td>Ensembl</td>
- * <td>Uniprot</td>
- * </tr>
- * <tr align=center>
- * <td>peptides</td>
- * <td>Ensembl</td>
- * <td>Uniprot</td>
- * </tr>
- * <tr align=center>
- * <td>dna</td>
- * <td>Ensembl</td>
- * <td>ENA</td>
- * </tr>
- * </table>
- *
- * @param sequence
- */
- public static void ensurePrimaries(SequenceI sequence)
- {
- List<DBRefEntry> pr = sequence.getPrimaryDBRefs();
- if (pr.size() == 0)
- {
- // nothing to do
- return;
- }
- List<DBRefEntry> selfs = new ArrayList<>();
- {
- DBRefEntry[] selfArray = selectDbRefs(!sequence.isProtein(),
- sequence.getDBRefs());
- if (selfArray == null || selfArray.length == 0)
- {
- // nothing to do
- return;
- }
- selfs.addAll(Arrays.asList(selfArray));
- }
-
- // filter non-primary refs
- for (DBRefEntry p : pr)
- {
- while (selfs.contains(p))
- {
- selfs.remove(p);
- }
- }
- List<DBRefEntry> toPromote = new ArrayList<>();
-
- for (DBRefEntry p : pr)
- {
- List<String> promType = new ArrayList<>();
- if (sequence.isProtein())
- {
- switch (getCanonicalName(p.getSource()))
- {
- case DBRefSource.UNIPROT:
- // case DBRefSource.UNIPROTKB:
- // case DBRefSource.UP_NAME:
- // search for and promote ensembl
- promType.add(DBRefSource.ENSEMBL);
- break;
- case DBRefSource.ENSEMBL:
- // search for and promote Uniprot
- promType.add(DBRefSource.UNIPROT);
- break;
- }
- }
- else
- {
- // TODO: promote transcript refs
- }
-
- // collate candidates and promote them
- DBRefEntry[] candidates = selectRefs(selfs.toArray(new DBRefEntry[0]),
- promType.toArray(new String[0]));
- if (candidates != null)
- {
- for (DBRefEntry cand : candidates)
- {
- if (cand.hasMap())
- {
- if (cand.getMap().getTo() != null
- && cand.getMap().getTo() != sequence)
- {
- // can't promote refs with mappings to other sequences
- continue;
- }
- if (cand.getMap().getMap().getFromLowest() != sequence
- .getStart()
- && cand.getMap().getMap().getFromHighest() != sequence
- .getEnd())
- {
- // can't promote refs with mappings from a region of this sequence
- // - eg CDS
- continue;
- }
- }
- // and promote
- cand.setVersion(p.getVersion() + " (promoted)");
- selfs.remove(cand);
- toPromote.add(cand);
- if (!cand.isPrimaryCandidate())
- {
- System.out.println(
- "Warning: Couldn't promote dbref " + cand.toString()
- + " for sequence " + sequence.toString());
- }
- }
- }
- }
- }
+ /**
- * Parses a DBRefEntry and adds it to the sequence, also a PDBEntry if the
- * database is PDB.
- * <p>
- * Used by file parsers to generate DBRefs from annotation within file (eg
- * Stockholm)
- *
- * @param dbname
- * @param version
- * @param acn
- * @param seq where to annotate with reference
- * @return parsed version of entry that was added to seq (if any)
- */
- public static DBRefEntry parseToDbRef(SequenceI seq, String dbname, String version, String acn) {
- DBRefEntry ref = null;
- if (dbname != null) {
- String locsrc = DBRefUtils.getCanonicalName(dbname);
- if (locsrc.equals(DBRefSource.PDB)) {
- /*
- * Check for PFAM style stockhom PDB accession id citation e.g. "1WRI A; 7-80;"
- */
- Regex r = new com.stevesoft.pat.Regex("([0-9][0-9A-Za-z]{3})\\s*(.?)\\s*;\\s*([0-9]+)-([0-9]+)");
- if (r.search(acn.trim())) {
- String pdbid = r.stringMatched(1);
- String chaincode = r.stringMatched(2);
- if (chaincode == null) {
- chaincode = " ";
- }
- // String mapstart = r.stringMatched(3);
- // String mapend = r.stringMatched(4);
- if (chaincode.equals(" ")) {
- chaincode = "_";
- }
- // construct pdb ref.
- ref = new DBRefEntry(locsrc, version, pdbid + chaincode);
- PDBEntry pdbr = new PDBEntry();
- pdbr.setId(pdbid);
- pdbr.setType(PDBEntry.Type.PDB);
- pdbr.setChainCode(chaincode);
- seq.addPDBId(pdbr);
- } else {
- System.err.println("Malformed PDB DR line:" + acn);
- }
- } else {
- // default:
- ref = new DBRefEntry(locsrc, version, acn);
- }
- }
- if (ref != null) {
- seq.addDBRef(ref);
- }
- return ref;
- }
-
- /**
+ * Returns true if either object is null, or they are equal
+ *
+ * @param o1
+ * @param o2
+ * @return
+ */
+ public static boolean nullOrEqual(Object o1, Object o2) {
+ if (o1 == null || o2 == null) {
+ return true;
+ }
+ return o1.equals(o2);
+ }
+
+ /**
+ * canonicalise source string before comparing. null is always wildcard
+ *
+ * @param o1 - null or source string to compare
+ * @param o2 - null or source string to compare
+ * @return true if either o1 or o2 are null, or o1 equals o2 under
+ * DBRefUtils.getCanonicalName
+ * (o1).equals(DBRefUtils.getCanonicalName(o2))
+ */
+ public static boolean nullOrEqualSource(String o1, String o2) {
+ if (o1 == null || o2 == null) {
+ return true;
+ }
+ return DBRefUtils.getCanonicalName(o1).equals(DBRefUtils.getCanonicalName(o2));
+ }
+
+ /**
+ * Selects just the DNA or protein references from a set of references
+ *
+ * @param selectDna if true, select references to 'standard' DNA databases, else
+ * to 'standard' peptide databases
+ * @param refs a set of references to select from
+ * @return
+ */
+ public static List<DBRefEntry> selectDbRefs(boolean selectDna, List<DBRefEntry> refs) {
+ return selectRefs(refs, selectDna ? DBRefSource.DNACODINGDBS : DBRefSource.PROTEINDBS);
+ // could attempt to find other cross
+ // refs here - ie PDB xrefs
+ // (not dna, not protein seq)
+ }
+
+ /**
+ * Returns the (possibly empty) list of those supplied dbrefs which have the
+ * specified source database, with a case-insensitive match of source name
+ *
+ * @param dbRefs
+ * @param source
+ * @return
+ */
+ public static List<DBRefEntry> searchRefsForSource(List<DBRefEntry> dbRefs, String source) {
+ List<DBRefEntry> matches = new ArrayList<DBRefEntry>();
+ if (dbRefs != null && source != null) {
+ for (DBRefEntry dbref : dbRefs) {
+ if (source.equalsIgnoreCase(dbref.getSource())) {
+ matches.add(dbref);
+ }
+ }
+ }
+ return matches;
+ }
+
+ /**
+ * promote direct database references to primary for nucleotide or protein
+ * sequences if they have an appropriate primary ref
+ * <table>
+ * <tr>
+ * <th>Seq Type</th>
+ * <th>Primary DB</th>
+ * <th>Direct which will be promoted</th>
+ * </tr>
+ * <tr align=center>
+ * <td>peptides</td>
+ * <td>Ensembl</td>
+ * <td>Uniprot</td>
+ * </tr>
+ * <tr align=center>
+ * <td>peptides</td>
+ * <td>Ensembl</td>
+ * <td>Uniprot</td>
+ * </tr>
+ * <tr align=center>
+ * <td>dna</td>
+ * <td>Ensembl</td>
+ * <td>ENA</td>
+ * </tr>
+ * </table>
+ *
+ * @param sequence
+ */
+ public static void ensurePrimaries(SequenceI sequence, List<DBRefEntry> pr) {
+ if (pr.size() == 0) {
+ // nothing to do
+ return;
+ }
+ int sstart = sequence.getStart();
+ int send = sequence.getEnd();
+ boolean isProtein = sequence.isProtein();
+ BitSet bsSelect = new BitSet();
+
+ // List<DBRefEntry> selfs = new ArrayList<DBRefEntry>();
+ // {
+
+ // List<DBRefEntry> selddfs = selectDbRefs(!isprot, sequence.getDBRefs());
+ // if (selfs == null || selfs.size() == 0)
+ // {
+ // // nothing to do
+ // return;
+ // }
+
+ List<DBRefEntry> dbrefs = sequence.getDBRefs();
+ bsSelect.set(0, dbrefs.size());
+
+ if (!selectRefsBS(dbrefs, isProtein ? DBRefSource.PROTEIN_MASK : DBRefSource.DNA_CODING_MASK, bsSelect))
+ return;
+
+ // selfs.addAll(selfArray);
+ // }
+
+ // filter non-primary refs
+ for (int ip = pr.size(); --ip >= 0;) {
+ DBRefEntry p = pr.get(ip);
+ for (int i = bsSelect.nextSetBit(0); i >= 0; i = bsSelect.nextSetBit(i + 1)) {
+ if (dbrefs.get(i) == p)
+ bsSelect.clear(i);
+ }
+ // while (selfs.contains(p))
+ // {
+ // selfs.remove(p);
+ // }
+ }
+ // List<DBRefEntry> toPromote = new ArrayList<DBRefEntry>();
+
+ for (int ip = pr.size(), keys = 0; --ip >= 0 && keys != DBRefSource.PRIMARY_MASK;) {
+ DBRefEntry p = pr.get(ip);
+ if (isProtein) {
+ switch (getCanonicalName(p.getSource())) {
+ case DBRefSource.UNIPROT:
+ keys |= DBRefSource.UNIPROT_MASK;
+ break;
+ case DBRefSource.ENSEMBL:
+ keys |= DBRefSource.ENSEMBL_MASK;
+ break;
+ }
+ } else {
+ // TODO: promote transcript refs ??
+ }
+ if (keys == 0 || !selectRefsBS(dbrefs, keys, bsSelect))
+ return;
+ // if (candidates != null)
+ {
+ for (int ic = bsSelect.nextSetBit(0); ic >= 0; ic = bsSelect.nextSetBit(ic + 1))
+ // for (int ic = 0, n = candidates.size(); ic < n; ic++)
+ {
+ DBRefEntry cand = dbrefs.get(ic);// candidates.get(ic);
+ if (cand.hasMap()) {
+ Mapping map = cand.getMap();
+ SequenceI cto = map.getTo();
+ if (cto != null && cto != sequence) {
+ // can't promote refs with mappings to other sequences
+ continue;
+ }
+ MapList mlist = map.getMap();
+ if (mlist.getFromLowest() != sstart && mlist.getFromHighest() != send) {
+ // can't promote refs with mappings from a region of this sequence
+ // - eg CDS
+ continue;
+ }
+ }
+ // and promote - not that version must be non-null here,
+ // as p must have passed isPrimaryCandidate()
+ cand.setVersion(p.getVersion() + " (promoted)");
+ bsSelect.clear(ic);
+ // selfs.remove(cand);
+ // toPromote.add(cand);
+ if (!cand.isPrimaryCandidate()) {
+ System.out.println("Warning: Couldn't promote dbref " + cand.toString() + " for sequence "
+ + sequence.toString());
+ }
+ }
+ }
+ }
+ }
}
git://source.jalview.org / jalview.git / commitdiff