- /**
- * Calls the Jmol library to parse the PDB file, and then inspects the
- * resulting object model to generate Jalview-style sequences, with secondary
- * structure annotation added where available (i.e. where it has been computed
- * by Jmol using DSSP).
- *
- * @see jalview.io.AlignFile#parse()
- */
- @Override
- public void parse() throws IOException
- {
-
- setChains(new Vector<PDBChain>());
- Viewer jmolModel = getJmolData();
- jmolModel.openReader(getDataName(), getDataName(), getReader());
- waitForScript(jmolModel);
-
- /*
- * Convert one or more Jmol Model objects to Jalview sequences
- */
- if (jmolModel.ms.mc > 0)
- {
- // parseBiopolymer(jmolModel.ms);
- transformJmolModelToJalview(jmolModel.ms);
- }
- }
-
- /**
- * Process the Jmol BioPolymer array and generate a Jalview sequence for each
- * chain found (including any secondary structure annotation from DSSP)
- *
- * @param ms
- * @throws IOException
- */
- public void parseBiopolymer(ModelSet ms) throws IOException
- {
- int modelIndex = -1;
- for (Model model : ms.am)
- {
- modelIndex++;
- String modelTitle = (String) ms.getInfo(modelIndex, "title");
- /*
- * Chains can span BioPolymers, so first make a flattened list, and then
- * work out the lengths of chains present
- */
- List<Monomer> monomers = getMonomers(ms, (BioModel) model);
- List<Integer> chainLengths = getChainLengths(monomers);
-
- /*
- * now chop up the Monomer list to make Jalview Sequences
- */
- int from = 0;
- for (int length : chainLengths)
- {
- buildSequenceFromChain(monomers.subList(from, from + length),
- modelTitle);
- from += length;
- }
- }
- }
-
- public void transformJmolModelToJalview(ModelSet ms)
- {
- try
- {
- String lastID = "";
- List<SequenceI> rna = new ArrayList<SequenceI>();
- List<SequenceI> prot = new ArrayList<SequenceI>();
- PDBChain tmpchain;
- String pdbId = (String) ms.getInfo(0, "title");
- setId(pdbId);
- List<Atom> significantAtoms = convertSignificantAtoms(ms);
- for (Atom tmpatom : significantAtoms)
- {
- try
- {
- tmpchain = findChain(tmpatom.chain);
- if (tmpatom.resNumIns.trim().equals(lastID))
- {
- // phosphorylated protein - seen both CA and P..
- continue;
- }
- tmpchain.atoms.addElement(tmpatom);
- } catch (Exception e)
- {
- tmpchain = new PDBChain(pdbId, tmpatom.chain);
- getChains().add(tmpchain);
- tmpchain.atoms.addElement(tmpatom);
- }
- lastID = tmpatom.resNumIns.trim();
- }
- makeResidueList();
- makeCaBondList();
-
- if (getId() == null)
- {
- setId(inFile.getName());
- }
- for (PDBChain chain : getChains())
- {
- SequenceI chainseq = postProcessChain(chain);
- if (isRNA(chainseq))
- {
- rna.add(chainseq);
- }
- else
- {
- prot.add(chainseq);
- }
- }
- } catch (OutOfMemoryError er)
- {
- System.out
- .println("OUT OF MEMORY LOADING TRANSFORMING JMOL MODEL TO JALVIEW MODEL");
- // throw new IOException(
- // MessageManager
- // .getString("exception.outofmemory_loading_pdb_file"));
- }
- }
-
- private List<Atom> convertSignificantAtoms(ModelSet ms)
- {
- List<Atom> significantAtoms = new ArrayList<Atom>();
- for (org.jmol.modelset.Atom atom : ms.at)
- {
- if (atom.getAtomName().equalsIgnoreCase("CA")
- || atom.getAtomName().equalsIgnoreCase("P"))
- {
- Atom curAtom = new Atom(atom.x, atom.y, atom.z);
- curAtom.atomIndex = atom.getIndex();
- curAtom.chain = atom.getChainIDStr();
- curAtom.insCode = atom.group.getInsertionCode();
- curAtom.name = atom.getAtomName();
- curAtom.number = atom.getAtomNumber();
- curAtom.resName = atom.getGroup3(true);
- curAtom.resNumber = atom.getResno();
- curAtom.occupancy = ms.occupancies != null ? ms.occupancies[atom
- .getIndex()] : Float.valueOf(atom.getOccupancy100());
- curAtom.resNumIns = "" + curAtom.resNumber + curAtom.insCode;
- curAtom.tfactor = 0;
- curAtom.type = 0;
- significantAtoms.add(curAtom);
- }
- }
- return significantAtoms;
- }
-
- /**
- * Helper method to construct a sequence for one chain and add it to the seqs
- * list
- *
- * @param monomers
- * a list of all monomers in the chain
- * @param modelTitle
- */
- protected void buildSequenceFromChain(List<Monomer> monomers,
- String modelTitle)
- {
- final int length = monomers.size();
-
- /*
- * arrays to hold sequence and secondary structure
- */
- char[] seq = new char[length];
- char[] secstr = new char[length];
- char[] secstrcode = new char[length];
-
- /*
- * populate the sequence and secondary structure arrays
- */
- extractJmolChainData(monomers, seq, secstr, secstrcode);
-
- /*
- * grab chain code and start position from first residue;
- */
- String chainId = monomers.get(0).chain.getIDStr();
- int firstResNum = monomers.get(0).getResno();
- if (firstResNum < 1)
- {
- // Jalview doesn't like residue < 1, so force this to 1
- System.err.println("Converting chain " + chainId + " first RESNUM ("
- + firstResNum + ") to 1");
- firstResNum = 1;
- }
-
- /*
- * convert any non-gap unknown residues to 'X'
- */
- convertNonGapCharacters(seq);
-
- /*
- * construct and add the Jalview sequence
- */
- String seqName = "" + modelTitle + "|" + chainId;
- int start = firstResNum;
- int end = firstResNum + length - 1;
-
- SequenceI sq = new Sequence(seqName, seq, start, end);
-
- addPdbid(sq, modelTitle, chainId);
-
- addSourceDBref(sq, modelTitle, start, end);
-
- seqs.add(sq);
-
- /*
- * add secondary structure predictions (if any)
- */
- if (isPredictSecondaryStructure())
- {
- addSecondaryStructureAnnotation(modelTitle, sq, secstr, secstrcode,
- chainId, firstResNum);
- }
-
- }
-
- /**
- * Add a source db ref entry for the given sequence.
- *
- * @param sq
- * @param accessionId
- * @param start
- * @param end
- */
- protected void addSourceDBref(SequenceI sq, String accessionId,
- int start, int end)
- {
- DBRefEntry sourceDBRef = new DBRefEntry();
- sourceDBRef.setAccessionId(accessionId);
- sourceDBRef.setSource(DBRefSource.MMCIF);
- sourceDBRef.setStartRes(start);
- sourceDBRef.setEndRes(end);
- sq.setSourceDBRef(sourceDBRef);
- sq.addDBRef(sourceDBRef);
- }
-
- /**
- * Add a PDBEntry giving the source of PDB data to the sequence
- *
- * @param sq
- * @param id
- * @param chainId
- */
- protected void addPdbid(SequenceI sq, String id, String chainId)
- {
- PDBEntry entry = new PDBEntry();
- entry.setId(id);
- entry.setType(PDBEntry.Type.MMCIF);
- entry.setProperty(new Hashtable());
- if (chainId != null)
- {
- // entry.getProperty().put("CHAIN", chains.elementAt(i).id);
- entry.setChainCode(String.valueOf(chainId));
- }
- if (inFile != null)
- {
- entry.setFile(inFile.getAbsolutePath());
- }
- else
- {
- // TODO: decide if we should dump the datasource to disk
- entry.setFile(getDataName());
- }
-
- sq.addPDBId(entry);
- }
-
- /**
- * Scans the list of (Jmol) Monomer objects, and adds the residue for each to
- * the sequence array, and any converted secondary structure prediction to the
- * secondary structure arrays
- *
- * @param monomers
- * @param seq
- * @param secstr
- * @param secstrcode
- */
- protected void extractJmolChainData(List<Monomer> monomers, char[] seq,
- char[] secstr, char[] secstrcode)
- {
- int pos = 0;
- for (Monomer monomer : monomers)
- {
- seq[pos] = monomer.getGroup1();
-
- /*
- * JAL-1828 replace a modified amino acid with its standard equivalent
- * (e.g. MSE with MET->M) to maximise sequence matching
- */
- replaceNonCanonicalResidue(monomer.getGroup3(), seq, pos);
-
- /*
- * if Jmol has derived a secondary structure prediction for this position,
- * convert it to Jalview equivalent and save it
- */
- setSecondaryStructure(monomer.getProteinStructureSubType(), pos,
- secstr, secstrcode);
- pos++;
- }
- }
-
- /**
- * Helper method that adds an AlignmentAnnotation for secondary structure to
- * the sequence, provided at least one secondary structure prediction has been
- * made
- *
- * @param modelTitle
- * @param seq
- * @param secstr
- * @param secstrcode
- * @param chainId
- * @param firstResNum
- * @return
- */
- protected void addSecondaryStructureAnnotation(String modelTitle,
- SequenceI sq, char[] secstr, char[] secstrcode, String chainId,
- int firstResNum)
- {
- char[] seq = sq.getSequence();
- boolean ssFound = false;
- Annotation asecstr[] = new Annotation[seq.length + firstResNum - 1];
- for (int p = 0; p < seq.length; p++)
- {
- if (secstr[p] >= 'A' && secstr[p] <= 'z')
- {
- asecstr[p] = new Annotation(String.valueOf(secstr[p]), null,
- secstrcode[p], Float.NaN);
- ssFound = true;
- }
- }
-
- if (ssFound)
- {
- String mt = modelTitle == null ? getDataName() : modelTitle;
- mt += chainId;
- AlignmentAnnotation ann = new AlignmentAnnotation(
- "Secondary Structure", "Secondary Structure for " + mt,
- asecstr);
- ann.belowAlignment = true;
- ann.visible = true;
- ann.autoCalculated = false;
- ann.setCalcId(getClass().getName());
- ann.adjustForAlignment();
- ann.validateRangeAndDisplay();
- annotations.add(ann);
- sq.addAlignmentAnnotation(ann);
- }
- }
-
- /**
- * Replace any non-gap miscellaneous characters with 'X'
- *
- * @param seq
- * @return
- */
- protected void convertNonGapCharacters(char[] seq)
- {
- boolean isNa = Comparison.areNucleotide(new char[][] { seq });
- int[] cinds = isNa ? ResidueProperties.nucleotideIndex
- : ResidueProperties.aaIndex;
- int nonGap = isNa ? ResidueProperties.maxNucleotideIndex
- : ResidueProperties.maxProteinIndex;
-
- for (int p = 0; p < seq.length; p++)
- {
- if (cinds[seq[p]] == nonGap)
- {
- seq[p] = 'X';
- }
- }
- }
-
- /**
- * Scans the list of Monomers (residue models), inspecting the chain id for
- * each, and returns an array whose length is the number of chains, and values
- * the length of each chain
- *
- * @param monomers
- * @return
- */
- protected List<Integer> getChainLengths(List<Monomer> monomers)