-/*\r
- * Jalview - A Sequence Alignment Editor and Viewer (Version 2.7)\r
- * Copyright (C) 2011 J Procter, AM Waterhouse, J Engelhardt, LM Lui, G Barton, M Clamp, S Searle\r
- * \r
- * This file is part of Jalview.\r
- * \r
- * Jalview is free software: you can redistribute it and/or\r
- * modify it under the terms of the GNU General Public License \r
- * as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.\r
- * \r
- * Jalview is distributed in the hope that it will be useful, but \r
- * WITHOUT ANY WARRANTY; without even the implied warranty \r
- * of MERCHANTABILITY or FITNESS FOR A PARTICULAR \r
- * PURPOSE. See the GNU General Public License for more details.\r
- * \r
- * You should have received a copy of the GNU General Public License along with Jalview. If not, see <http://www.gnu.org/licenses/>.\r
- */\r
-package jalview.datamodel;\r
-\r
-/**\r
- * Defines internal constants for unambiguous annotation of DbRefEntry source\r
- * strings and describing the data retrieved from external database sources (see\r
- * jalview.ws.DbSourcProxy)\r
- * \r
- * @author JimP\r
- * \r
- */\r
-public class DBRefSource\r
-{\r
- /**\r
- * UNIPROT Accession Number\r
- */\r
- public static String UNIPROT = "UNIPROT";\r
-\r
- /**\r
- * UNIPROT Entry Name\r
- */\r
- public static String UP_NAME = "UNIPROT_NAME";\r
-\r
- /**\r
- * Uniprot Knowledgebase/TrEMBL as served from EMBL protein products.\r
- */\r
- public static final String UNIPROTKB = "UniProtKB/TrEMBL";\r
-\r
- /**\r
- * PDB Entry Code\r
- */\r
- public static String PDB = "PDB";\r
-\r
- /**\r
- * EMBL ID\r
- */\r
- public static String EMBL = "EMBL";\r
-\r
- /**\r
- * EMBLCDS ID\r
- */\r
- public static String EMBLCDS = "EMBLCDS";\r
-\r
- /**\r
- * PFAM ID\r
- */\r
- public static String PFAM = "PFAM";\r
- \r
- /**\r
- * RFAM ID\r
- */\r
- public static String RFAM = "RFAM";\r
-\r
- /**\r
- * GeneDB ID\r
- */\r
- public static final String GENEDB = "GeneDB";\r
-\r
- /**\r
- * List of databases whose sequences might have coding regions annotated\r
- */\r
- public static final String[] DNACODINGDBS =\r
- { EMBL, EMBLCDS, GENEDB };\r
-\r
- public static final String[] CODINGDBS =\r
- { EMBLCDS, GENEDB };\r
-\r
- public static final String[] PROTEINDBS =\r
- { UNIPROT, PDB, UNIPROTKB };\r
-\r
- public static final String[] PROTEINSEQ =\r
- { UNIPROT, UNIPROTKB };\r
-\r
- public static final String[] PROTEINSTR =\r
- { PDB };\r
-\r
- public static final String[] DOMAINDBS =\r
- { PFAM, RFAM };\r
-\r
- /**\r
- * set of unique DBRefSource property constants. These could be used to\r
- * reconstruct the above groupings\r
- */\r
- public static final Object SEQDB = "SQ";\r
-\r
- /**\r
- * database of nucleic acid sequences\r
- */\r
- public static final Object DNASEQDB = "NASQ";\r
-\r
- /**\r
- * database of amino acid sequences\r
- */\r
- public static final Object PROTSEQDB = "PROTSQ";\r
-\r
- /**\r
- * database of cDNA sequences\r
- */\r
- public static final Object CODINGSEQDB = "CODING";\r
-\r
- /**\r
- * database of na sequences with exon annotation\r
- */\r
- public static final Object DNACODINGSEQDB = "XONCODING";\r
-\r
- /**\r
- * DB returns several sequences associated with a protein/nucleotide domain\r
- */\r
- public static final Object DOMAINDB = "DOMAIN";\r
-\r
- /**\r
- * DB query can take multiple accession codes concatenated by a separator.\r
- * Value of property indicates maximum number of accession codes to send at a\r
- * time.\r
- */\r
- public static final Object MULTIACC = "MULTIACC";\r
-\r
- /**\r
- * DB query returns an alignment for each accession provided.\r
- */\r
- public static final Object ALIGNMENTDB = "ALIGNMENTS";\r
-}\r
+/*
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
+ *
+ * This file is part of Jalview.
+ *
+ * Jalview is free software: you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, either version 3
+ * of the License, or (at your option) any later version.
+ *
+ * Jalview is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ * PURPOSE. See the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
+ */
+package jalview.datamodel;
+
+/**
+ * BH 2018 SwingJS note: If additional final static Strings are added to this
+ * file, they should be added to public static final String[] allTypes.
+ *
+ * Defines internal constants for unambiguous annotation of DbRefEntry source
+ * strings and describing the data retrieved from external database sources (see
+ * jalview.ws.DbSourcProxy) <br/>
+ * TODO: replace with ontology to allow recognition of particular attributes
+ * (e.g. protein coding, alignment (ortholog db, paralog db, domain db),
+ * genomic, transcriptomic, 3D structure providing (PDB, MODBASE, etc) ..).
+ *
+ *
+ *
+ * @author JimP
+ *
+ */
+public class DBRefSource
+{
+
+
+
+ public static final String UNIPROT = "UNIPROT";
+ 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 ENSEMBL = "ENSEMBL";
+ public static final String ENSEMBLGENOMES = "ENSEMBLGENOMES";
+
+ public static final String EMBL = "EMBL";
+ public static final String EMBLCDS = "EMBLCDS";
+ public static final String EMBLCDSProduct = "EMBLCDSProtein".toUpperCase();
+
+ 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();
+
+ 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[] CODINGDBS = { EMBLCDS, GENEDB, ENSEMBL };
+
+ public static final int CODING_MASK = EMBL_CDS_MASK | GENE_DB_MASK | ENSEMBL_MASK;
+
+
+
+ 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);
+ }
+
+
+// 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 (ucversion.startsWith(allSources[i])) // BH 2019.01.25 .toUpperCase() unnecessary here for allSources
+ {
+ // 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 true;
+}
+
+
+
+
+}
git://source.jalview.org / jalview.git / blobdiff