X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FCrossRef.java;h=41582b4598679f45424d667d06c96aef75cec477;hb=f411449b59b0a7369f6f77cd1a78b7417e36ab50;hp=c54357e316baad23bbaaccc8ce6eced00d527be3;hpb=0b573ed90b14079f7326281f50c0c9cffdace586;p=jalview.git diff --git a/src/jalview/analysis/CrossRef.java b/src/jalview/analysis/CrossRef.java index c54357e..41582b4 100644 --- a/src/jalview/analysis/CrossRef.java +++ b/src/jalview/analysis/CrossRef.java @@ -143,14 +143,16 @@ public class CrossRef /* * first find seq's xrefs (dna-to-peptide or peptide-to-dna) */ - List rfs = DBRefUtils.selectDbRefs(!fromDna, seq.getDBRefs()); + List rfs = DBRefUtils.selectDbRefs(!fromDna, + seq.getDBRefs()); addXrefsToSources(rfs, sources); if (dataset != null) { /* * find sequence's direct (dna-to-dna, peptide-to-peptide) xrefs */ - List lrfs = DBRefUtils.selectDbRefs(fromDna, seq.getDBRefs()); + List lrfs = DBRefUtils.selectDbRefs(fromDna, + seq.getDBRefs()); List foundSeqs = new ArrayList<>(); /* @@ -291,7 +293,7 @@ public class CrossRef if (matchInDataset != null && xref.getMap().getTo() != null && matchInDataset != xref.getMap().getTo()) { - System.err.println( + jalview.bin.Console.errPrintln( "Implementation problem (reopen JAL-2154): CrossRef.findInDataset seems to have recovered a different sequence than the one explicitly mapped for xref." + "Found:" + matchInDataset + "\nExpected:" + xref.getMap().getTo() + "\nFor xref:" @@ -370,7 +372,8 @@ public class CrossRef { // 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) { @@ -421,7 +424,7 @@ public class CrossRef retrieved = sftch.getSequences(sourceRefs, !fromDna); } catch (Exception e) { - System.err.println( + jalview.bin.Console.errPrintln( "Problem whilst retrieving cross references for Sequence : " + seq.getName()); e.printStackTrace(); @@ -443,6 +446,11 @@ public class CrossRef 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 @@ -493,11 +501,12 @@ public class CrossRef // protein if (sq.isProtein() == fromDna) { - List sqdbrefs = sq.getPrimaryDBRefs(); + List sqdbrefs = sq.getPrimaryDBRefs(); for (int idb = 0, ndb = sqdbrefs.size(); idb < ndb; idb++) { - DBRefEntry dbr = sqdbrefs.get(idb); - List searchrefs = DBRefUtils.searchRefs(dbrSourceSet, dbr, DBRefUtils.SEARCH_MODE_FULL); + DBRefEntry dbr = sqdbrefs.get(idb); + List searchrefs = DBRefUtils.searchRefs(dbrSourceSet, + dbr, DBRefUtils.SEARCH_MODE_FULL); for (int isr = 0, nsr = searchrefs.size(); isr < nsr; isr++) { sourceRefs.remove(searchrefs.get(isr)); @@ -516,7 +525,9 @@ public class CrossRef /** * 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 @@ -535,10 +546,11 @@ public class CrossRef List 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) { @@ -550,7 +562,7 @@ public class CrossRef Mapping map = dbref.getMap(); if (map != null) { - SequenceI ms = map.getTo(); + SequenceI ms = map.getTo(); if (ms != null && map.getMap() != null) { if (ms == sourceSequence) @@ -595,7 +607,7 @@ public class CrossRef String msg = "Mapping updated from " + ms.getName() + " to retrieved crossreference " + matched.getName(); - System.out.println(msg); + jalview.bin.Console.outPrintln(msg); List toRefs = map.getTo().getDBRefs(); if (toRefs != null) @@ -650,7 +662,7 @@ public class CrossRef cf.addMap(retrievedSequence, map.getTo(), map.getMap()); } catch (Exception e) { - System.err.println( + jalview.bin.Console.errPrintln( "Exception when consolidating Mapped sequence set..."); e.printStackTrace(System.err); } @@ -715,7 +727,8 @@ public class CrossRef /** * 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 @@ -746,8 +759,8 @@ public class CrossRef for (SequenceI seq : dataset.getSequences()) { // first check primary refs. - List match = DBRefUtils.searchRefs( - seq.getPrimaryDBRefs(), template, DBRefUtils.SEARCH_MODE_FULL); + List match = DBRefUtils.searchRefs(seq.getPrimaryDBRefs(), + template, DBRefUtils.SEARCH_MODE_FULL); if (match != null && match.size() == 1 && sameSequence(seq, dss)) { return seq; @@ -813,7 +826,8 @@ public class CrossRef /** * 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 @@ -964,10 +978,10 @@ public class CrossRef } 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); } @@ -1000,7 +1014,8 @@ public class CrossRef * sequenceI or all the returned sequences (eg a genomic reference * associated with a locus and one or more transcripts) * - * @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, @@ -1014,7 +1029,7 @@ public class CrossRef } if (dataset.getSequences() == null) { - System.err.println("Empty dataset sequence set - NO VECTOR"); + jalview.bin.Console.errPrintln("Empty dataset sequence set - NO VECTOR"); return false; } List ds = dataset.getSequences(); @@ -1026,7 +1041,7 @@ public class CrossRef { if (nxt.getDatasetSequence() != null) { - System.err.println( + jalview.bin.Console.errPrintln( "Implementation warning: CrossRef initialised with a dataset alignment with non-dataset sequences in it! (" + nxt.getDisplayId(true) + " has ds reference " + nxt.getDatasetSequence().getDisplayId(true)