*/
package jalview.datamodel;
-import java.lang.reflect.Field;
-import java.util.ArrayList;
-import java.util.List;
-
/**
* BH 2018 SwingJS note: If additional final static Strings are added to this
* file, they should be added to public static final String[] allTypes.
- /**
- * UNIPROT Accession Number
- */
public static final String UNIPROT = "UNIPROT";
-
- /**
- * UNIPROT Entry Name
- */
public static final String UP_NAME = "UNIPROT_NAME".toUpperCase();
-
/**
* Uniprot Knowledgebase/TrEMBL as served from EMBL protein products.
*/
public static final String UNIPROTKB = "UniProtKB/TrEMBL".toUpperCase();
- public static final String EMBLCDSProduct = "EMBLCDSProtein".toUpperCase();
-
-
- /**
- * PDB Entry Code
- */
- public static final String PDB = "PDB";
-
- /**
- * EMBL ID
- */
- public static final String EMBL = "EMBL";
-
- /**
- * EMBLCDS ID
- */
- public static final String EMBLCDS = "EMBLCDS";
-
+ public static final String ENSEMBL = "ENSEMBL";
+ public static final String ENSEMBLGENOMES = "ENSEMBLGENOMES";
- /**
- * PFAM ID
- */
- public static final String PFAM = "PFAM";
-
- /**
- * RFAM ID
- */
- public static final String RFAM = "RFAM";
+ public static final String EMBL = "EMBL";
+ public static final String EMBLCDS = "EMBLCDS";
+ public static final String EMBLCDSProduct = "EMBLCDSProtein".toUpperCase();
- /**
- * GeneDB ID
- */
+ public static final String PDB = "PDB";
+ public static final String PFAM = "PFAM";
+ public static final String RFAM = "RFAM";
public static final String GENEDB = "GeneDB".toUpperCase();
-
- /**
- * Ensembl
- */
- public static final String ENSEMBL = "ENSEMBL";
+ public static final String PDB_CANONICAL_NAME = PDB;
+
+
+ public static final String[] allSources = new String[] {
+ UNIPROT,
+ UP_NAME, UNIPROTKB,
+ ENSEMBL, ENSEMBLGENOMES,
+ EMBL, EMBLCDS, EMBLCDSProduct,
+ PDB, PFAM, RFAM, GENEDB
+ };
+
+ public static final int UNIPROT_MASK = 1<<0;
+ public static final int UP_NAME_MASK = 1<<1;
+ public static final int UNIPROT_KB_MASK = 1<<2;
+ public static final int ENSEMBL_MASK = 1<<3;
+ public static final int ENSEMBL_GENOMES_MASK = 1<<4;
+ public static final int EMBL_MASK = 1<<5;
+ public static final int EMBL_CDS_MASK = 1<<6;
+ public static final int EMBL_CDS_PRODUCT_MASK = 1<<7;
+ public static final int PDB_MASK = 1<<8;
+ public static final int PFAM_MASK = 1<<9;
+ public static final int RFAM_MASK = 1<<10;
+ public static final int GENE_DB_MASK = 1<<11;
+
+ public static final int MASK_COUNT = 12;
+
+ public static final int ALL_MASKS = (1 << MASK_COUNT) - 1;
+
+ public static int getSourceKey(String name) {
+ for (int i = 0; i < MASK_COUNT; i++) {
+ if (name.equals(allSources[i]))
+ return 1<<i;
+ }
+ return 0;
+ }
+
+ public static final int PRIMARY_MASK = UNIPROT_MASK | ENSEMBL_MASK;
+
+ /**
+ * List of databases whose sequences might have coding regions annotated
+ */
+ public static final String[] DNACODINGDBS = {
+ ENSEMBL, ENSEMBLGENOMES,
+ EMBL, EMBLCDS, GENEDB
+ };
+
+ public static final int DNA_CODING_MASK =
+ ENSEMBL_MASK | ENSEMBL_GENOMES_MASK
+ | EMBL_MASK | EMBL_CDS_MASK | GENE_DB_MASK;
- public static final String ENSEMBLGENOMES = "ENSEMBLGENOMES";
+
+
+ public static final String[] CODINGDBS = { EMBLCDS, GENEDB, ENSEMBL };
+ public static final int CODING_MASK = EMBL_CDS_MASK | GENE_DB_MASK | ENSEMBL_MASK;
- /**
- * List of databases whose sequences might have coding regions annotated
- */
- public static final String[] DNACODINGDBS = { EMBL, EMBLCDS, GENEDB,
- ENSEMBL, ENSEMBLGENOMES };
-
- public static final String[] CODINGDBS = { EMBLCDS, GENEDB, ENSEMBL };
-
- public static final String[] PROTEINDBS = { UNIPROT, UNIPROTKB,
- EMBLCDSProduct, ENSEMBL }; // Ensembl ENSP* entries are protein
-
+
- public static final String[] allTypes = new String[] {
- UNIPROT, UP_NAME, UNIPROTKB,
- EMBLCDSProduct, PDB, EMBL,
- EMBLCDS, PFAM, RFAM,
- GENEDB, ENSEMBL, ENSEMBLGENOMES
- };
-
-public static final String PROTEINDBSKEYS, DNACODINGDBSKEYS;
-
-public static final String[] PROMTYPES;
-
-
-public static final int UNIPROT_MASK = 1;
-
-public static final int ENSEMBL_MASK = 2;
-
-public static final int ALL_MASKS = UNIPROT_MASK | ENSEMBL_MASK;
-
-public static final String PDB_CANONICAL_NAME = PDB;
-
-static {
- // BH 2019.01.25 trying to speed this up
- String s = ";";
- for (int i = PROTEINDBS.length; --i >= 0;)
- s += PROTEINDBS[i] + ";";
- PROTEINDBSKEYS = s;
-
- s = ";";
- for (int i = DNACODINGDBS.length; --i >= 0;)
- s += DNACODINGDBS[i] + ";";
- DNACODINGDBSKEYS = s;
-
- PROMTYPES = new String[] { null, ";" + UNIPROT + ";", ";" + ENSEMBL + ";" , ";" + UNIPROT + ";" + ENSEMBL + ";" };
-}
-
- public static String[] allSourcesFromReflection;
-
- public static String[] allSources()
+ public static final String[] PROTEINDBS = {
+ UNIPROT, UNIPROTKB,
+ ENSEMBL, EMBLCDSProduct }; // Ensembl ENSP* entries are protein
+
+ public static final int PROTEIN_MASK =
+ UNIPROT_MASK | UNIPROT_KB_MASK
+ | ENSEMBL_MASK | EMBL_CDS_PRODUCT_MASK ;
+
+
+ // for SequenceAnnotationReport only
+
+// public static final String[][] PRIMARY_SOURCES = new String[][] {
+// CODINGDBS, DNACODINGDBS, PROTEINDBS };
+//
+ public static final int PRIMARY_SOURCES_MASK = CODING_MASK | DNA_CODING_MASK | PROTEIN_MASK;
+
+ public static boolean isPrimarySource(String source)
+ {
+ return ((PRIMARY_SOURCES_MASK & getSourceKey(source)) != 0);
+ }
- {
- /**
- * @j2sNative
- *
- * return C$.allTypes;
- *
- */
+// public static String[] allSourcesFromReflection;
+//
+// public static String[] allSources()
+//
+// {
+// /**
+// * @j2sNative
+// *
+// * return C$.allTypes;
+// *
+// */
+//
+// {
+// if (allSourcesFromReflection == null)
+// {
+// List<String> src = new ArrayList<>();
+// for (Field f : DBRefSource.class.getFields())
+// {
+// if (String.class.equals(f.getType()))
+// {
+// try
+// {
+// src.add((String) f.get(null));
+// } catch (Exception x)
+// {
+// x.printStackTrace();
+// }
+// }
+// }
+// allSourcesFromReflection = src.toArray(new String[0]);
+// }
+// return allSourcesFromReflection;
+// }
+// }
+
+ public static boolean isPrimaryCandidate(String ucversion) {
+ // tricky - this test really needs to search the sequence's set of dbrefs to
+ // see if there is a primary reference that derived this reference.
+ for (int i = allSources.length; --i >= 0;)
{
- if (allSourcesFromReflection == null)
+ if (ucversion.startsWith(allSources[i])) // BH 2019.01.25 .toUpperCase() unnecessary here for allSources
{
- List<String> src = new ArrayList<>();
- for (Field f : DBRefSource.class.getFields())
- {
- if (String.class.equals(f.getType()))
- {
- try
- {
- src.add((String) f.get(null));
- } catch (Exception x)
- {
- x.printStackTrace();
- }
- }
- }
- allSourcesFromReflection = src.toArray(new String[0]);
+ // by convention, many secondary references inherit the primary
+ // reference's
+ // source string as a prefix for any version information from the
+ // secondary reference.
+ return false;
}
- return allSourcesFromReflection;
}
- }
-
+ return true;
+}
+
+
+
}
/**
* Utilities for handling DBRef objects and their collections.
*/
-public class DBRefUtils
-{
-
- public final static int DB_SOURCE = 1;
+public class DBRefUtils {
+
+ public final static int DB_SOURCE = 1;
public final static int DB_VERSION = 2;
- public final static int DB_ID = 4;
- public final static int DB_MAP = 8;
-
-
+ public final static int DB_ID = 4;
+ public final static int DB_MAP = 8;
+
public final static int SEARCH_MODE_NO_MAP_NO_VERSION = DB_SOURCE | DB_ID;
public final static int SEARCH_MODE_FULL = DB_SOURCE | DB_VERSION | DB_ID | DB_MAP;
- /*
- * lookup from lower-case form of a name to its canonical (standardised) form
- */
- private static Map<String, String> canonicalSourceNameLookup = new HashMap<String, String>();
+ /*
+ * lookup from lower-case form of a name to its canonical (standardised) form
+ */
+ private static Map<String, String> canonicalSourceNameLookup = new HashMap<String, String>();
- private static Map<String, String> dasCoordinateSystemsLookup = new HashMap<String, String>();
+ private static Map<String, String> dasCoordinateSystemsLookup = new HashMap<String, String>();
- static
- {
- // TODO load these from a resource file?
- canonicalSourceNameLookup.put("uniprotkb/swiss-prot",
- DBRefSource.UNIPROT);
- canonicalSourceNameLookup.put("uniprotkb/trembl", DBRefSource.UNIPROT);
+ static {
+ // TODO load these from a resource file?
+ canonicalSourceNameLookup.put("uniprotkb/swiss-prot", DBRefSource.UNIPROT);
+ canonicalSourceNameLookup.put("uniprotkb/trembl", DBRefSource.UNIPROT);
- // Ensembl values for dbname in xref REST service:
- canonicalSourceNameLookup.put("uniprot/sptrembl", DBRefSource.UNIPROT);
- canonicalSourceNameLookup.put("uniprot/swissprot", DBRefSource.UNIPROT);
+ // Ensembl values for dbname in xref REST service:
+ canonicalSourceNameLookup.put("uniprot/sptrembl", DBRefSource.UNIPROT);
+ canonicalSourceNameLookup.put("uniprot/swissprot", DBRefSource.UNIPROT);
- canonicalSourceNameLookup.put("pdb", DBRefSource.PDB);
- canonicalSourceNameLookup.put("ensembl", DBRefSource.ENSEMBL);
- // Ensembl Gn and Tr are for Ensembl genomic and transcript IDs as served
- // from ENA.
- canonicalSourceNameLookup.put("ensembl-tr", DBRefSource.ENSEMBL);
- canonicalSourceNameLookup.put("ensembl-gn", DBRefSource.ENSEMBL);
+ canonicalSourceNameLookup.put("pdb", DBRefSource.PDB);
+ canonicalSourceNameLookup.put("ensembl", DBRefSource.ENSEMBL);
+ // Ensembl Gn and Tr are for Ensembl genomic and transcript IDs as served
+ // from ENA.
+ canonicalSourceNameLookup.put("ensembl-tr", DBRefSource.ENSEMBL);
+ canonicalSourceNameLookup.put("ensembl-gn", DBRefSource.ENSEMBL);
- // Make sure we have lowercase entries for all canonical string lookups
+ // Make sure we have lowercase entries for all canonical string lookups
// BH 2019.01.25 unnecessary -- they are all lower case already
- //Set<String> keys = canonicalSourceNameLookup.keySet();
+ // Set<String> keys = canonicalSourceNameLookup.keySet();
// for (String k : keys)
// {
// canonicalSourceNameLookup.put(k.toLowerCase(),
// canonicalSourceNameLookup.get(k));
// }
- dasCoordinateSystemsLookup.put("pdbresnum", DBRefSource.PDB);
- dasCoordinateSystemsLookup.put("uniprot", DBRefSource.UNIPROT);
- dasCoordinateSystemsLookup.put("embl", DBRefSource.EMBL);
- // dasCoordinateSystemsLookup.put("embl", DBRefSource.EMBLCDS);
- }
-
- /**
- * Returns those DBRefEntry objects whose source identifier (once converted to
- * Jalview's canonical form) is in the list of sources to search for. Returns
- * null if no matches found.
- *
- * @param dbrefs
- * DBRefEntry objects to search
- * @param sources
- * array of sources to select
- * @return
- */
- public static List<DBRefEntry> selectRefs(List<DBRefEntry> dbrefs,
- String[] sources)
- {
- if (dbrefs == null || sources == null)
- {
- return dbrefs;
- }
-
- // BH TODO
- HashSet<String> srcs = new HashSet<String>();
- for (String src : sources)
- {
- srcs.add(src.toUpperCase());
- }
-
- int nrefs = dbrefs.size();
- List<DBRefEntry> res = new ArrayList<DBRefEntry>();
- for (int ib = 0; ib < nrefs; ib++)
- {
- DBRefEntry dbr = dbrefs.get(ib);
- String source = getCanonicalName(dbr.getSource());
- if (srcs.contains(source.toUpperCase()))
- {
- res.add(dbr);
- }
- }
-
- if (res.size() > 0)
- {
- //List<DBRefEntry> reply = new DBRefEntry[res.size()];
- return res;//.toArray(reply);
- }
- return null;
- }
-
- private static boolean selectRefsBS(List<DBRefEntry> dbrefs, String sourceKeys, BitSet bsSelect) {
- if (dbrefs == null || sourceKeys == null)
- {
- return false;
- }
- for (int i = 0, n = dbrefs.size(); i < n; i++)
- {
- DBRefEntry dbr = dbrefs.get(i);
- String sourceKey = dbr.getSourceKey();
- if (sourceKeys.indexOf(sourceKey) < 0) {
- bsSelect.clear(i);
- }
- }
- return !bsSelect.isEmpty();
- }
-
-
- /**
- * isDasCoordinateSystem
- *
- * @param string
- * String
- * @param dBRefEntry
- * DBRefEntry
- * @return boolean true if Source DBRefEntry is compatible with DAS
- * CoordinateSystem name
- */
-
- public static boolean isDasCoordinateSystem(String string,
- DBRefEntry dBRefEntry)
- {
- if (string == null || dBRefEntry == null)
- {
- return false;
- }
- String coordsys = dasCoordinateSystemsLookup.get(string.toLowerCase());
- return coordsys == null ? false
- : coordsys.equals(dBRefEntry.getSource());
- }
-
- /**
- * 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;
+ dasCoordinateSystemsLookup.put("pdbresnum", DBRefSource.PDB);
+ dasCoordinateSystemsLookup.put("uniprot", DBRefSource.UNIPROT);
+ dasCoordinateSystemsLookup.put("embl", DBRefSource.EMBL);
+ // dasCoordinateSystemsLookup.put("embl", DBRefSource.EMBLCDS);
+ }
+
+ /**
+ * Returns those DBRefEntry objects whose source identifier (once converted to
+ * Jalview's canonical form) is in the list of sources to search for. Returns
+ * null if no matches found.
+ *
+ * @param dbrefs DBRefEntry objects to search
+ * @param sources array of sources to select
+ * @return
+ */
+ public static List<DBRefEntry> selectRefs(List<DBRefEntry> dbrefs, String[] sources) {
+ if (dbrefs == null || sources == null) {
+ return dbrefs;
+ }
+
+ // BH TODO
+ HashSet<String> srcs = new HashSet<String>();
+ for (String src : sources) {
+ srcs.add(src.toUpperCase());
+ }
+
+ int nrefs = dbrefs.size();
+ List<DBRefEntry> res = new ArrayList<DBRefEntry>();
+ for (int ib = 0; ib < nrefs; ib++) {
+ DBRefEntry dbr = dbrefs.get(ib);
+ String source = getCanonicalName(dbr.getSource());
+ if (srcs.contains(source.toUpperCase())) {
+ res.add(dbr);
+ }
+ }
+
+ if (res.size() > 0) {
+ // List<DBRefEntry> reply = new DBRefEntry[res.size()];
+ return res;// .toArray(reply);
+ }
+ return null;
+ }
+
+ private static boolean selectRefsBS(List<DBRefEntry> dbrefs, int sourceKeys, BitSet bsSelect) {
+ if (dbrefs == null || sourceKeys == 0) {
+ return false;
+ }
+ for (int i = 0, n = dbrefs.size(); i < n; i++) {
+ DBRefEntry dbr = dbrefs.get(i);
+ if ((dbr.getSourceKey() & sourceKeys) != 0) {
+ bsSelect.clear(i);
+ }
+ }
+ return !bsSelect.isEmpty();
+ }
+
+ /**
+ * isDasCoordinateSystem
+ *
+ * @param string String
+ * @param dBRefEntry DBRefEntry
+ * @return boolean true if Source DBRefEntry is compatible with DAS
+ * CoordinateSystem name
+ */
+
+ public static boolean isDasCoordinateSystem(String string, DBRefEntry dBRefEntry) {
+ if (string == null || dBRefEntry == null) {
+ return false;
+ }
+ String coordsys = dasCoordinateSystemsLookup.get(string.toLowerCase());
+ return coordsys == null ? false : coordsys.equals(dBRefEntry.getSource());
+ }
+
+ /**
+ * 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);
-}
+ }
+
+ /**
+ * 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;
+ }
+
+ };
+
+ /**
+ * 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;
+ }
+ };
+
+ /**
+ * accession ID only must be identical.
+ */
+ public static DbRefComp matchId = new DbRefComp() {
+ @Override
+ public boolean matches(DBRefEntry refa, DBRefEntry refb, int mode) {
+ if (refa.getAccessionId() != null && refb.getAccessionId() != null
+ && refb.getAccessionId().equals(refa.getAccessionId())) {
+ 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);
+ }
+ }
+ 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();
- /**
- * 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, 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;
- }
-
- };
-
- /**
- * 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;
- }
- };
-
- /**
- * accession ID only must be identical.
- */
- public static DbRefComp matchId = new DbRefComp()
- {
- @Override
- public boolean matches(DBRefEntry refa, DBRefEntry refb, int mode)
- {
- if (refa.getAccessionId() != null && refb.getAccessionId() != null
- && refb.getAccessionId().equals(refa.getAccessionId()))
- {
- 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);
- }
- }
- 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>();
// {
// return;
// }
- List<DBRefEntry> dbrefs = sequence.getDBRefs();
- bsSelect.set(0, dbrefs.size());
-
- if (!selectRefsBS(dbrefs, isProtein ? DBRefSource.PROTEINDBSKEYS : DBRefSource.DNACODINGDBSKEYS, bsSelect))
- 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);
- }
+ // 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.ALL_MASKS;)
- {
- 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, DBRefSource.PROMTYPES[keys], bsSelect))
- return;
+ 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 = 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
- cand.setVersion(p.getVersion() + " (promoted)");
- bsSelect.clear(ic);
- //selfs.remove(cand);
+ {
+ 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());
- }
- }
- }
- }
- }
+ if (!cand.isPrimaryCandidate()) {
+ System.out.println("Warning: Couldn't promote dbref " + cand.toString() + " for sequence "
+ + sequence.toString());
+ }
+ }
+ }
+ }
+ }
}