From: Jim Procter Date: Sun, 11 Feb 2018 11:03:27 +0000 (+0000) Subject: JAL-2780 JAL-2781 use PDBe numbering to map between sequence extracted from PDB chain... X-Git-Tag: Release_2_10_4~68^2~20 X-Git-Url: http://source.jalview.org/gitweb/?a=commitdiff_plain;h=38889ac4815d29d1fbb040ff1dbbc47fe89ea87e;p=jalview.git JAL-2780 JAL-2781 use PDBe numbering to map between sequence extracted from PDB chain and Uniprot mapping --- diff --git a/src/jalview/ws/sifts/SiftsClient.java b/src/jalview/ws/sifts/SiftsClient.java index e1b539e..3899f04 100644 --- a/src/jalview/ws/sifts/SiftsClient.java +++ b/src/jalview/ws/sifts/SiftsClient.java @@ -108,6 +108,8 @@ public class SiftsClient implements SiftsClientI private static final int PDB_ATOM_POS = 1; + private static final int PDBE_POS = 2; + private static final String NOT_OBSERVED = "Not_Observed"; private static final String SIFTS_FTP_BASE_URL = "http://ftp.ebi.ac.uk/pub/databases/msd/sifts/xml/"; @@ -458,8 +460,8 @@ public class SiftsClient implements SiftsClientI public HashMap getGreedyMapping(String entityId, SequenceI seq, java.io.PrintStream os) throws SiftsException { - List omitNonObserved = new ArrayList(); - int nonObservedShiftIndex = 0; + List omitNonObserved = new ArrayList<>(); + int nonObservedShiftIndex = 0,pdbeNonObserved=0; // System.out.println("Generating mappings for : " + entityId); Entity entity = null; entity = getEntityById(entityId); @@ -490,7 +492,7 @@ public class SiftsClient implements SiftsClientI TreeMap resNumMap = new TreeMap(); List segments = entity.getSegment(); SegmentHelperPojo shp = new SegmentHelperPojo(seq, mapping, resNumMap, - omitNonObserved, nonObservedShiftIndex); + omitNonObserved, nonObservedShiftIndex,pdbeNonObserved); processSegments(segments, shp); try { @@ -628,6 +630,7 @@ public class SiftsClient implements SiftsClientI List residues = segment.getListResidue().getResidue(); for (Residue residue : residues) { + boolean isObserved = isResidueObserved(residue); int pdbeIndex = getLeadingIntegerValue(residue.getDbResNum(), UNASSIGNED); int currSeqIndex = UNASSIGNED; @@ -638,6 +641,19 @@ public class SiftsClient implements SiftsClientI if (cRefDb.getDbSource().equalsIgnoreCase(DBRefSource.PDB)) { pdbRefDb = cRefDb; + if (firstPDBResNum == UNASSIGNED) + { + firstPDBResNum = getLeadingIntegerValue(cRefDb.getDbResNum(), + UNASSIGNED); + } + else + { + if (isObserved) + { + // after we find the first observed residue we just increment + firstPDBResNum++; + } + } } if (cRefDb.getDbCoordSys().equalsIgnoreCase(seqCoordSys.getName()) && isAccessionMatched(cRefDb.getDbAccessionId())) @@ -650,6 +666,10 @@ public class SiftsClient implements SiftsClientI } } } + if (!isObserved) + { + ++pdbeNonObservedCount; + } if (seqCoordSys == seqCoordSys.PDB) // FIXME: is seqCoordSys ever PDBe // ??? { @@ -662,15 +682,18 @@ public class SiftsClient implements SiftsClientI { continue; } - if (!isResidueObserved(residue) - && seqCoordSys != seqCoordSys.UNIPROT) + if (!isObserved) { - // mapping to PDB or PDBe so we need to bookkeep for the non-observed - // SEQRES positions - omitNonObserved.add(currSeqIndex); - ++nonObservedShiftIndex; + if (seqCoordSys != CoordinateSys.UNIPROT) // FIXME: PDB or PDBe only + // here + { + // mapping to PDB or PDBe so we need to bookkeep for the + // non-observed + // SEQRES positions + omitNonObserved.add(currSeqIndex); + ++nonObservedShiftIndex; + } } - // if (currSeqIndex >= seq.getStart() && currSeqIndex <= seqlength) // // true // numbering @@ -695,10 +718,13 @@ public class SiftsClient implements SiftsClientI .getCanonicalAminoAcid(residue.getDbResName())); resNumMap.put(currSeqIndex, String.valueOf(resCharCode)); } - + // TODO: mapping should have SeqCoordSysNum => int[] {PDBeNum, + // PDBRESNUM, ATOMNUM } mapping.put(currSeqIndex - nonObservedShiftIndex, new int[] - { Integer.valueOf(resNum), UNASSIGNED }); + { Integer.valueOf(resNum), UNASSIGNED, + isObserved ? firstPDBResNum + : UNASSIGNED }); } } } @@ -988,17 +1014,36 @@ public class SiftsClient implements SiftsClientI private int nonObservedShiftIndex; + /** + * count of number of 'not observed' positions in the PDB record's SEQRES + * (total number of residues with coordinates == length(SEQRES) - + * pdbeNonObserved + */ + private int pdbeNonObserved; + public SegmentHelperPojo(SequenceI seq, HashMap mapping, TreeMap resNumMap, - List omitNonObserved, int nonObservedShiftIndex) + List omitNonObserved, int nonObservedShiftIndex, + int pdbeNonObserved) { setSeq(seq); setMapping(mapping); setResNumMap(resNumMap); setOmitNonObserved(omitNonObserved); setNonObservedShiftIndex(nonObservedShiftIndex); + setPdbeNonObserved(pdbeNonObserved); + } + public void setPdbeNonObserved(int pdbeNonObserved2) + { + this.pdbeNonObserved = pdbeNonObserved2; + } + + public int getPdbeNonObserved() + { + return pdbeNonObserved; + } public SequenceI getSeq() { return seq; @@ -1048,6 +1093,7 @@ public class SiftsClient implements SiftsClientI { this.nonObservedShiftIndex = nonObservedShiftIndex; } + } @Override