*/
package jalview.analysis;
+import java.util.ArrayList;
+import java.util.Iterator;
+import java.util.List;
+
import jalview.datamodel.AlignedCodonFrame;
import jalview.datamodel.Alignment;
import jalview.datamodel.AlignmentI;
import jalview.datamodel.SequenceI;
import jalview.util.DBRefUtils;
import jalview.util.MapList;
-import jalview.ws.SequenceFetcherFactory;
-import jalview.ws.seqfetcher.ASequenceFetcher;
-
-import java.util.ArrayList;
-import java.util.Iterator;
-import java.util.List;
+import jalview.ws.SequenceFetcher;
/**
* Functions for cross-referencing sequence databases.
/*
* first find seq's xrefs (dna-to-peptide or peptide-to-dna)
*/
- List<DBRefEntry> rfs = DBRefUtils.selectDbRefs(!fromDna, seq.getDBRefs());
+ List<DBRefEntry> rfs = DBRefUtils.selectDbRefs(!fromDna,
+ seq.getDBRefs());
addXrefsToSources(rfs, sources);
if (dataset != null)
{
/*
* find sequence's direct (dna-to-dna, peptide-to-peptide) xrefs
*/
- List<DBRefEntry> lrfs = DBRefUtils.selectDbRefs(fromDna, seq.getDBRefs());
+ List<DBRefEntry> lrfs = DBRefUtils.selectDbRefs(fromDna,
+ seq.getDBRefs());
List<SequenceI> foundSeqs = new ArrayList<>();
/*
{
// do a bit more work - search for sequences with references matching
// xrefs on this sequence.
- found = searchDataset(fromDna, dss, xref, rseqs, cf, false, DBRefUtils.SEARCH_MODE_FULL);
+ found = searchDataset(fromDna, dss, xref, rseqs, cf, false,
+ DBRefUtils.SEARCH_MODE_FULL);
}
if (found)
{
private void retrieveCrossRef(List<DBRefEntry> sourceRefs, SequenceI seq,
List<DBRefEntry> xrfs, boolean fromDna, AlignedCodonFrame cf)
{
- ASequenceFetcher sftch = SequenceFetcherFactory.getSequenceFetcher();
SequenceI[] retrieved = null;
SequenceI dss = seq.getDatasetSequence() == null ? seq
: seq.getDatasetSequence();
}
try
{
- retrieved = sftch.getSequences(sourceRefs, !fromDna);
+ retrieved = SequenceFetcher.getInstance().getSequences(sourceRefs, !fromDna);
} catch (Exception e)
{
System.err.println(
addedXref |= importCrossRefSeq(cf, newDsSeqs, doNotAdd, dss,
retrievedDss);
}
+ // JBPNote: What assumptions are made for dbref structures on
+ // retrieved sequences ?
+ // addedXref will be true means importCrossRefSeq found
+ // sequences with dbrefs with mappings to sequences congruent with dss
+
if (!addedXref)
{
// try again, after looking for matching IDs
private void removeAlreadyRetrievedSeqs(List<DBRefEntry> sourceRefs,
boolean fromDna)
{
- List<DBRefEntry> dbrSourceSet = new ArrayList<DBRefEntry>(sourceRefs);
+ List<DBRefEntry> dbrSourceSet = new ArrayList<>(sourceRefs);
List<SequenceI> dsSeqs = dataset.getSequences();
- for (int ids = dsSeqs.size(); --ids >= 0;)
+ for (int ids = 0, nds = dsSeqs.size(); ids < nds; ids++)
{
SequenceI sq = dsSeqs.get(ids);
boolean dupeFound = false;
// protein
if (sq.isProtein() == fromDna)
{
- List<DBRefEntry> sqdbrefs = sq.getPrimaryDBRefs();
- for (int idb = sqdbrefs.size(); --idb >= 0;)
+ List<DBRefEntry> sqdbrefs = sq.getPrimaryDBRefs();
+ for (int idb = 0, ndb = sqdbrefs.size(); idb < ndb; idb++)
{
- DBRefEntry dbr = sqdbrefs.get(idb);
- List<DBRefEntry> searchrefs = DBRefUtils.searchRefs(dbrSourceSet, dbr, DBRefUtils.SEARCH_MODE_FULL);
- for (int isr = searchrefs.size(); --isr >= 0;)
+ DBRefEntry dbr = sqdbrefs.get(idb);
+ List<DBRefEntry> searchrefs = DBRefUtils.searchRefs(dbrSourceSet,
+ dbr, DBRefUtils.SEARCH_MODE_FULL);
+ for (int isr = 0, nsr = searchrefs.size(); isr < nsr; isr++)
{
sourceRefs.remove(searchrefs.get(isr));
dupeFound = true;
/**
* process sequence retrieved via a dbref on source sequence to resolve and
- * transfer data
+ * transfer data JBPNote: as of 2022-02-03 - this assumes retrievedSequence
+ * has dbRefs with Mapping references to a sequence congruent with
+ * sourceSequence
*
* @param cf
* @param sourceSequence
List<DBRefEntry> dbr = retrievedSequence.getDBRefs();
if (dbr != null)
{
- for (int ib = 0, nb = dbr.size(); ib < nb; ib++)
+ for (int ib = 0, nb = dbr.size(); ib < nb; ib++)
{
- DBRefEntry dbref = dbr.get(ib);
+ DBRefEntry dbref = dbr.get(ib);
+ // matched will return null if the dbref has no map
SequenceI matched = findInDataset(dbref);
if (matched == sourceSequence)
{
Mapping map = dbref.getMap();
if (map != null)
{
- SequenceI ms = map.getTo();
+ SequenceI ms = map.getTo();
if (ms != null && map.getMap() != null)
{
if (ms == sourceSequence)
/**
* Returns null or the first sequence in the dataset which is identical to
* xref.mapTo, and has a) a primary dbref matching xref, or if none found, the
- * first one with an ID source|xrefacc
+ * first one with an ID source|xrefacc JBPNote: Could refactor this to
+ * AlignmentI/DatasetI
*
* @param xref
* with map and mapped-to sequence
for (SequenceI seq : dataset.getSequences())
{
// first check primary refs.
- List<DBRefEntry> match = DBRefUtils.searchRefs(
- seq.getPrimaryDBRefs(), template, DBRefUtils.SEARCH_MODE_FULL);
+ List<DBRefEntry> match = DBRefUtils.searchRefs(seq.getPrimaryDBRefs(),
+ template, DBRefUtils.SEARCH_MODE_FULL);
if (match != null && match.size() == 1 && sameSequence(seq, dss))
{
return seq;
/**
* Updates any empty mappings in the cross-references with one to a compatible
* retrieved sequence if found, and adds any new mappings to the
- * AlignedCodonFrame
+ * AlignedCodonFrame JBPNote: TODO: this relies on sequence IDs like
+ * UNIPROT|ACCESSION - which do not always happen.
*
* @param mapFrom
* @param xrefs
if (fromDna)
{
- AlignmentUtils.computeProteinFeatures(mapFrom, mapTo, mapping);
+ // AlignmentUtils.computeProteinFeatures(mapFrom, mapTo, mapping);
mappings.addMap(mapFrom, mapTo, mapping);
}
else
}
for (int i = 0, n = lrfs.size(); i < n; i++)
{
-// DBRefEntry xref = new DBRefEntry(lrfs.get(i));
-// // add in wildcards
-// xref.setVersion(null);
-// xref.setMap(null);
+ // DBRefEntry xref = new DBRefEntry(lrfs.get(i));
+ // // add in wildcards
+ // xref.setVersion(null);
+ // xref.setMap(null);
found |= searchDataset(fromDna, sequenceI, lrfs.get(i), foundSeqs, cf,
false, DBRefUtils.SEARCH_MODE_NO_MAP_NO_VERSION);
}
* sequenceI or all the returned sequences (eg a genomic reference
* associated with a locus and one or more transcripts)</li>
* </ul>
- * @param mode SEARCH_MODE_FULL for all; SEARCH_MODE_NO_MAP_NO_VERSION optional
+ * @param mode
+ * SEARCH_MODE_FULL for all; SEARCH_MODE_NO_MAP_NO_VERSION optional
* @return true if relationship found and sequence added.
*/
boolean searchDataset(boolean fromDna, SequenceI fromSeq, DBRefEntry xrf,
git://source.jalview.org / jalview.git / blobdiff