+/*\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.analysis;\r
\r
import java.util.Enumeration;\r
+import java.util.List;\r
import java.util.Vector;\r
import java.util.Hashtable;\r
\r
import jalview.datamodel.DBRefEntry;\r
import jalview.datamodel.Sequence;\r
import jalview.datamodel.SequenceI;\r
-import jalview.ws.ASequenceFetcher;\r
import jalview.ws.SequenceFetcher;\r
+import jalview.ws.seqfetcher.ASequenceFetcher;\r
\r
/**\r
* Functions for cross-referencing sequence databases. user must first specify\r
* if cross-referencing from protein or dna (set dna==true)\r
- * \r
+ *\r
* @author JimP\r
- * \r
+ *\r
*/\r
public class CrossRef\r
{\r
/**\r
* get the DNA or protein references for a protein or dna sequence\r
- * \r
+ *\r
* @param dna\r
* @param rfs\r
* @return\r
{\r
rfs = jalview.util.DBRefUtils.selectRefs(rfs,\r
DBRefSource.DNACODINGDBS); // could attempt to find other cross\r
- // refs and return here - ie PDB xrefs\r
- // (not dna, not protein seq)\r
+ // refs and return here - ie PDB xrefs\r
+ // (not dna, not protein seq)\r
}\r
return rfs;\r
}\r
public static Hashtable classifyDbRefs(DBRefEntry[] rfs)\r
{\r
Hashtable classes = new Hashtable();\r
- classes.put(DBRefSource.PROTEINDBS, jalview.util.DBRefUtils.selectRefs(\r
- rfs, DBRefSource.PROTEINDBS));\r
+ classes.put(DBRefSource.PROTEINDBS,\r
+ jalview.util.DBRefUtils.selectRefs(rfs, DBRefSource.PROTEINDBS));\r
classes.put(DBRefSource.DNACODINGDBS, jalview.util.DBRefUtils\r
.selectRefs(rfs, DBRefSource.DNACODINGDBS));\r
- classes.put(DBRefSource.DOMAINDBS, jalview.util.DBRefUtils.selectRefs(\r
- rfs, DBRefSource.DOMAINDBS));\r
+ classes.put(DBRefSource.DOMAINDBS,\r
+ jalview.util.DBRefUtils.selectRefs(rfs, DBRefSource.DOMAINDBS));\r
// classes.put(OTHER, )\r
return classes;\r
}\r
\r
/**\r
* @param dna\r
- * true if seqs are DNA seqs\r
+ * true if seqs are DNA seqs\r
* @param seqs\r
* @return a list of sequence database cross reference source types\r
*/\r
/**\r
* Indirect references are references from other sequences from the dataset to\r
* any of the direct DBRefEntrys on the given sequences.\r
- * \r
+ *\r
* @param dna\r
- * true if seqs are DNA seqs\r
+ * true if seqs are DNA seqs\r
* @param seqs\r
* @return a list of sequence database cross reference source types\r
*/\r
Vector refs = new Vector();\r
for (int s = 0; s < seqs.length; s++)\r
{\r
- SequenceI dss = seqs[s];\r
- while (dss.getDatasetSequence() != null)\r
+ if (seqs[s] != null)\r
{\r
- dss = dss.getDatasetSequence();\r
- }\r
- DBRefEntry[] rfs = findXDbRefs(dna, dss.getDBRef());\r
- for (int r = 0; rfs != null && r < rfs.length; r++)\r
- {\r
- if (!refs.contains(rfs[r].getSource()))\r
+\r
+ SequenceI dss = seqs[s];\r
+ while (dss.getDatasetSequence() != null)\r
{\r
- refs.addElement(rfs[r].getSource());\r
+ dss = dss.getDatasetSequence();\r
}\r
- }\r
- if (dataset != null)\r
- {\r
- // search for references to this sequence's direct references.\r
- DBRefEntry[] lrfs = CrossRef.findXDbRefs(!dna, seqs[s].getDBRef());\r
- Vector rseqs = new Vector();\r
- CrossRef.searchDatasetXrefs(seqs[s], !dna, lrfs, dataset, rseqs,\r
- null); // don't need to specify codon frame for mapping here\r
- Enumeration lr = rseqs.elements();\r
- while (lr.hasMoreElements())\r
+ DBRefEntry[] rfs = findXDbRefs(dna, dss.getDBRef());\r
+ for (int r = 0; rfs != null && r < rfs.length; r++)\r
{\r
- SequenceI rs = (SequenceI) lr.nextElement();\r
- DBRefEntry[] xrs = findXDbRefs(dna, rs.getDBRef());\r
- for (int r = 0; rfs != null && r < rfs.length; r++)\r
+ if (!refs.contains(rfs[r].getSource()))\r
{\r
- if (!refs.contains(rfs[r].getSource()))\r
+ refs.addElement(rfs[r].getSource());\r
+ }\r
+ }\r
+ if (dataset != null)\r
+ {\r
+ // search for references to this sequence's direct references.\r
+ DBRefEntry[] lrfs = CrossRef\r
+ .findXDbRefs(!dna, seqs[s].getDBRef());\r
+ Vector rseqs = new Vector();\r
+ CrossRef.searchDatasetXrefs(seqs[s], !dna, lrfs, dataset, rseqs,\r
+ null); // don't need to specify codon frame for mapping here\r
+ Enumeration lr = rseqs.elements();\r
+ while (lr.hasMoreElements())\r
+ {\r
+ SequenceI rs = (SequenceI) lr.nextElement();\r
+ DBRefEntry[] xrs = findXDbRefs(dna, rs.getDBRef());\r
+ for (int r = 0; rfs != null && r < rfs.length; r++)\r
{\r
- refs.addElement(rfs[r].getSource());\r
+ if (!refs.contains(rfs[r].getSource()))\r
+ {\r
+ refs.addElement(rfs[r].getSource());\r
+ }\r
}\r
}\r
}\r
{\r
if (cdna[c].getSource().equals(DBRefSource.EMBLCDS))\r
{\r
- // retrieve CDS dataset sequences\r
+ System.err\r
+ .println("TODO: unimplemented sequence retrieval for coding region sequence.");\r
+ // TODO: retrieve CDS dataset sequences\r
// need global dataset sequence retriever/resolver to reuse refs\r
// and construct Mapping entry.\r
// insert gaps in CDS according to peptide gaps.\r
}\r
\r
/**\r
- * \r
+ *\r
* @param dna\r
* @param seqs\r
* @return\r
}\r
\r
/**\r
- * \r
+ *\r
* @param seqs\r
* @param dna\r
* @param source\r
* @param dataset\r
- * alignment to search for product sequences.\r
+ * alignment to search for product sequences.\r
* @return products (as dataset sequences)\r
*/\r
public static Alignment findXrefSequences(SequenceI[] seqs, boolean dna,\r
{\r
System.out.println("Attempting to find ds Xrefs refs.");\r
DBRefEntry[] lrfs = CrossRef.findXDbRefs(!dna, seqs[s].getDBRef()); // less\r
- // ambiguous\r
- // would\r
- // be a\r
- // 'find\r
- // primary\r
- // dbRefEntry'\r
- // method.\r
+ // ambiguous\r
+ // would\r
+ // be a\r
+ // 'find\r
+ // primary\r
+ // dbRefEntry'\r
+ // method.\r
// filter for desired source xref here\r
found = CrossRef.searchDatasetXrefs(dss, !dna, lrfs, dataset,\r
rseqs, cf);\r
// xrefs on this sequence.\r
if (dataset != null)\r
{\r
- found |= searchDataset(dss, xrfs[r], dataset, rseqs, cf);\r
+ found |= searchDataset(dss, xrfs[r], dataset, rseqs, cf); // ,false,!dna);\r
if (found)\r
xrfs[r] = null; // we've recovered seqs for this one.\r
}\r
xrfs = t;\r
try\r
{\r
- retrieved = sftch.getSequences(xrfs);\r
+ retrieved = sftch.getSequences(xrfs); // problem here is we don't\r
+ // know which of xrfs\r
+ // resulted in which\r
+ // retrieved element\r
} catch (Exception e)\r
{\r
System.err\r
{\r
for (int rs = 0; rs < retrieved.length; rs++)\r
{\r
+ // TODO: examine each sequence for 'redundancy'\r
+ jalview.datamodel.DBRefEntry[] dbr = retrieved[rs]\r
+ .getDBRef();\r
+ if (dbr != null && dbr.length > 0)\r
+ {\r
+ for (int di = 0; di < dbr.length; di++)\r
+ {\r
+ // find any entry where we should put in the sequence being\r
+ // cross-referenced into the map\r
+ jalview.datamodel.Mapping map = dbr[di].getMap();\r
+ if (map != null)\r
+ {\r
+ if (map.getTo() != null && map.getMap() != null)\r
+ {\r
+ // should search the local dataset to find any existing\r
+ // candidates for To !\r
+ try\r
+ {\r
+ // compare ms with dss and replace with dss in mapping\r
+ // if map is congruent\r
+ SequenceI ms = map.getTo();\r
+ int sf = map.getMap().getToLowest();\r
+ int st = map.getMap().getToHighest();\r
+ SequenceI mappedrg = ms.getSubSequence(sf, st);\r
+ SequenceI loc = dss.getSubSequence(sf, st);\r
+ if (mappedrg.getLength() > 0\r
+ && mappedrg.getSequenceAsString().equals(\r
+ loc.getSequenceAsString()))\r
+ {\r
+ System.err\r
+ .println("Mapping updated for retrieved crossreference");\r
+ // method to update all refs of existing To on\r
+ // retrieved sequence with dss and merge any props\r
+ // on To onto dss.\r
+ map.setTo(dss);\r
+ }\r
+ } catch (Exception e)\r
+ {\r
+ System.err\r
+ .println("Exception when consolidating Mapped sequence set...");\r
+ e.printStackTrace(System.err);\r
+ }\r
+ }\r
+ }\r
+ }\r
+ }\r
+ retrieved[rs].updatePDBIds();\r
rseqs.addElement(retrieved[rs]);\r
}\r
}\r
* find references to lrfs in the cross-reference set of each sequence in\r
* dataset (that is not equal to sequenceI) Identifies matching DBRefEntry\r
* based on source and accession string only - Map and Version are nulled.\r
- * \r
+ *\r
* @param sequenceI\r
* @param lrfs\r
* @param dataset\r
/**\r
* search a given sequence dataset for references matching cross-references to\r
* the given sequence\r
- * \r
+ *\r
* @param sequenceI\r
* @param xrf\r
* @param dataset\r
* @param rseqs\r
- * set of unique sequences\r
+ * set of unique sequences\r
* @param cf\r
* @return true if one or more unique sequences were found and added\r
*/\r
* TODO: generalise to different protein classifications Search dataset for\r
* DBRefEntrys matching the given one (xrf) and add the associated sequence to\r
* rseq.\r
- * \r
+ *\r
* @param sequenceI\r
* @param xrf\r
* @param dataset\r
* @param rseqs\r
- * @param direct -\r
- * search all references or only subset\r
+ * @param direct\r
+ * - search all references or only subset\r
* @param dna\r
- * search dna or protein xrefs (if direct=false)\r
+ * search dna or protein xrefs (if direct=false)\r
* @return true if relationship found and sequence added.\r
*/\r
public static boolean searchDataset(SequenceI sequenceI, DBRefEntry xrf,\r
boolean direct, boolean dna)\r
{\r
boolean found = false;\r
+ SequenceI[] typer = new SequenceI[1];\r
if (dataset == null)\r
return false;\r
if (dataset.getSequences() == null)\r
System.err.println("Empty dataset sequence set - NO VECTOR");\r
return false;\r
}\r
- Enumeration e = dataset.getSequences().elements();\r
- while (e.hasMoreElements())\r
+ List<SequenceI> ds;\r
+ synchronized (ds=dataset.getSequences())\r
{\r
- SequenceI nxt = (SequenceI) e.nextElement();\r
+ for (SequenceI nxt:ds)\r
if (nxt != null)\r
{\r
if (nxt.getDatasetSequence() != null)\r
}\r
if (nxt != sequenceI && nxt != sequenceI.getDatasetSequence())\r
{\r
+ // check if this is the correct sequence type\r
+ {\r
+ typer[0] = nxt;\r
+ boolean isDna = jalview.util.Comparison.isNucleotide(typer);\r
+ if ((direct && isDna == dna) || (!direct && isDna != dna))\r
+ {\r
+ // skip this sequence because it is same molecule type\r
+ continue;\r
+ }\r
+ }\r
+\r
// look for direct or indirect references in common\r
- DBRefEntry[] poss = null, cands = null;\r
+ DBRefEntry[] poss = nxt.getDBRef(), cands = null;\r
if (direct)\r
{\r
- cands = jalview.util.DBRefUtils.searchRefs(poss = nxt\r
- .getDBRef(), xrf);\r
+ cands = jalview.util.DBRefUtils.searchRefs(poss, xrf);\r
}\r
else\r
{\r
- cands = jalview.util.DBRefUtils.searchRefs(poss = CrossRef\r
- .findXDbRefs(dna, nxt.getDBRef()), xrf);\r
+ poss = CrossRef.findXDbRefs(dna, poss); //\r
+ cands = jalview.util.DBRefUtils.searchRefs(poss, xrf);\r
}\r
if (cands != null)\r
{\r
{\r
rseqs.addElement(nxt);\r
boolean foundmap = cf != null; // don't search if we aren't given\r
- // a codon map object\r
+ // a codon map object\r
for (int r = 0; foundmap && r < cands.length; r++)\r
{\r
if (cands[r].hasMap())\r
/**\r
* precalculate different products that can be found for seqs in dataset and\r
* return them.\r
- * \r
+ *\r
* @param dna\r
* @param seqs\r
* @param dataset\r
- * @param fake -\r
- * don't actually build lists - just get types\r
+ * @param fake\r
+ * - don't actually build lists - just get types\r
* @return public static Object[] buildXProductsList(boolean dna, SequenceI[]\r
* seqs, AlignmentI dataset, boolean fake) { String types[] =\r
* jalview.analysis.CrossRef.findSequenceXrefTypes( dna, seqs,\r
* System.out.println("Type: " + types[t]); SequenceI[] prod =\r
* jalview.analysis.CrossRef.findXrefSequences(seqs, dna, types[t]);\r
* System.out.println("Found " + ((prod == null) ? "no" : "" +\r
- * prod.length) + " products"); if (prod!=null) { for (int p=0; p<prod.length;\r
- * p++) { System.out.println("Prod "+p+":\r
- * "+prod[p].getDisplayId(true)); } } }\r
- * } else { System.out.println("Trying getProducts for\r
- * "+al.getSequenceAt(0).getDisplayId(true)); System.out.println("Search DS\r
- * Xref for: "+(dna ? "dna" : "prot")); // have a bash at finding the products\r
- * amongst all the retrieved sequences. SequenceI[] prod =\r
- * jalview.analysis.CrossRef.findXrefSequences(al .getSequencesArray(), dna,\r
- * null, ds); System.out.println("Found " + ((prod == null) ? "no" : "" +\r
- * prod.length) + " products"); if (prod!=null) { // select non-equivalent\r
- * sequences from dataset list for (int p=0; p<prod.length; p++) {\r
- * System.out.println("Prod "+p+": "+prod[p].getDisplayId(true)); } }\r
- * } }\r
+ * prod.length) + " products"); if (prod!=null) { for (int p=0;\r
+ * p<prod.length; p++) { System.out.println("Prod "+p+":\r
+ * "+prod[p].getDisplayId(true)); } } } } else {\r
+ * System.out.println("Trying getProducts for\r
+ * "+al.getSequenceAt(0).getDisplayId(true));\r
+ * System.out.println("Search DS Xref for: "+(dna ? "dna" : "prot"));\r
+ * // have a bash at finding the products amongst all the retrieved\r
+ * sequences. SequenceI[] prod =\r
+ * jalview.analysis.CrossRef.findXrefSequences(al\r
+ * .getSequencesArray(), dna, null, ds); System.out.println("Found " +\r
+ * ((prod == null) ? "no" : "" + prod.length) + " products"); if\r
+ * (prod!=null) { // select non-equivalent sequences from dataset list\r
+ * for (int p=0; p<prod.length; p++) { System.out.println("Prod "+p+":\r
+ * "+prod[p].getDisplayId(true)); } } } }\r
*/\r
-}
\ No newline at end of file
+}\r
git://source.jalview.org / jalview.git / blobdiff