+ return (PARSE_REGEX == null ? PARSE_REGEX = Platform.newRegex(
+ "([0-9][0-9A-Za-z]{3})\\s*(.?)\\s*;\\s*([0-9]+)-([0-9]+)", null)
+ : PARSE_REGEX);
+ }
+ /**
+ * 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 = getParseRegex();
+ 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<>();
+ 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))