import jalview.datamodel.AlignmentAnnotation;
import jalview.datamodel.Annotation;
-import jalview.datamodel.DBRefEntry;
-import jalview.datamodel.DBRefSource;
import jalview.datamodel.PDBEntry;
-import jalview.datamodel.Sequence;
import jalview.datamodel.SequenceI;
-import jalview.io.AlignFile;
import jalview.io.FileParse;
+import jalview.io.StructureFile;
import jalview.schemes.ResidueProperties;
-import jalview.util.Comparison;
+import jalview.structure.StructureImportSettings;
+import jalview.util.Format;
import jalview.util.MessageManager;
import java.io.IOException;
import java.util.ArrayList;
-import java.util.Collection;
-import java.util.Hashtable;
+import java.util.HashMap;
import java.util.List;
import java.util.Map;
-
-import javajs.awt.Dimension;
+import java.util.Vector;
import org.jmol.api.JmolStatusListener;
import org.jmol.api.JmolViewer;
import org.jmol.c.CBK;
import org.jmol.c.STR;
-import org.jmol.modelset.Group;
-import org.jmol.modelset.Model;
import org.jmol.modelset.ModelSet;
-import org.jmol.modelsetbio.BioModel;
-import org.jmol.modelsetbio.BioPolymer;
-import org.jmol.modelsetbio.Monomer;
import org.jmol.viewer.Viewer;
+import MCview.Atom;
import MCview.PDBChain;
+import MCview.Residue;
/**
- * Import and process files with Jmol for file like PDB, mmCIF
+ * Import and process files with Jmol for file like PDB, mmCIF
*
* @author jprocter
*
*/
-public class JmolParser extends AlignFile implements
- JmolStatusListener
+public class JmolParser extends StructureFile implements JmolStatusListener
{
Viewer viewer = null;
- private Collection<PDBChain> chains;
-
- /*
- * Set true to predict secondary structure (using JMol for protein, Annotate3D
- * for RNA)
- */
- private boolean predictSecondaryStructure = true;
-
public JmolParser(String inFile, String type) throws IOException
{
super(inFile, type);
}
/**
+ * Calls the Jmol library to parse the PDB/mmCIF 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)
+ {
+ // ideally we do this
+ // try
+ // {
+ // setStructureFileType(jmolModel.evalString("show _fileType"));
+ // } catch (Exception q)
+ // {
+ // }
+ // ;
+ // instead, we distinguish .cif from non-.cif by filename
+ setStructureFileType(getDataName().toLowerCase().endsWith(".cif") ? PDBEntry.Type.MMCIF
+ .toString() : "PDB");
+
+ transformJmolModelToJalview(jmolModel.ms);
+ }
+ }
+
+ /**
* create a headless jmol instance for dataprocessing
*
* @return
{
try
{
+ /*
+ * params -o (output to sysout) -n (nodisplay) -x (exit when finished)
+ * see http://wiki.jmol.org/index.php/Jmol_Application
+ */
viewer = (Viewer) JmolViewer.allocateViewer(null, null, null, null,
null, "-x -o -n", this);
// ensure the 'new' (DSSP) not 'old' (Ramachandran) SS method is used
return viewer;
}
+ public void transformJmolModelToJalview(ModelSet ms) throws IOException
+ {
+ try
+ {
+ String lastID = "";
+ List<SequenceI> rna = new ArrayList<SequenceI>();
+ List<SequenceI> prot = new ArrayList<SequenceI>();
+ PDBChain tmpchain;
+ String pdbId = (String) ms.getInfo(0, "title");
+
+ if (pdbId == null)
+ {
+ setId(safeName(getDataName()));
+ setPDBIdAvailable(false);
+ }
+ else
+ {
+ setId(pdbId);
+ setPDBIdAvailable(true);
+ }
+ 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(getId(), tmpatom.chain);
+ getChains().add(tmpchain);
+ tmpchain.atoms.addElement(tmpatom);
+ }
+ lastID = tmpatom.resNumIns.trim();
+ }
+ xferSettings();
+
+ makeResidueList();
+ makeCaBondList();
+
+ for (PDBChain chain : getChains())
+ {
+ SequenceI chainseq = postProcessChain(chain);
+ if (isRNA(chainseq))
+ {
+ rna.add(chainseq);
+ }
+ else
+ {
+ prot.add(chainseq);
+ }
+
+ if (StructureImportSettings.isProcessSecondaryStructure())
+ {
+ createAnnotation(chainseq, chain, ms.at);
+ }
+ }
+ } catch (OutOfMemoryError er)
+ {
+ System.out
+ .println("OUT OF MEMORY LOADING TRANSFORMING JMOL MODEL TO JALVIEW MODEL");
+ throw new IOException(
+ MessageManager
+ .getString("exception.outofmemory_loading_mmcif_file"));
+ }
+ }
+
+ private List<Atom> convertSignificantAtoms(ModelSet ms)
+ {
+ List<Atom> significantAtoms = new ArrayList<Atom>();
+ HashMap<String, org.jmol.modelset.Atom> chainTerMap = new HashMap<String, org.jmol.modelset.Atom>();
+ org.jmol.modelset.Atom prevAtom = null;
+ for (org.jmol.modelset.Atom atom : ms.at)
+ {
+ if (atom.getAtomName().equalsIgnoreCase("CA")
+ || atom.getAtomName().equalsIgnoreCase("P"))
+ {
+ if (!atomValidated(atom, prevAtom, chainTerMap))
+ {
+ continue;
+ }
+ Atom curAtom = new Atom(atom.x, atom.y, atom.z);
+ curAtom.atomIndex = atom.getIndex();
+ curAtom.chain = atom.getChainIDStr();
+ curAtom.insCode = atom.group.getInsertionCode() == '\000' ? ' '
+ : 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());
+ String fmt = new Format("%4i").form(curAtom.resNumber);
+ curAtom.resNumIns = (fmt + curAtom.insCode);
+ curAtom.tfactor = atom.getBfactor100() / 100f;
+ curAtom.type = 0;
+ // significantAtoms.add(curAtom);
+ // ignore atoms from subsequent models
+ if (!significantAtoms.contains(curAtom))
+ {
+ significantAtoms.add(curAtom);
+ }
+ prevAtom = atom;
+ }
+ }
+ return significantAtoms;
+ }
+
+ private boolean atomValidated(org.jmol.modelset.Atom curAtom,
+ org.jmol.modelset.Atom prevAtom,
+ HashMap<String, org.jmol.modelset.Atom> chainTerMap)
+ {
+ // System.out.println("Atom: " + curAtom.getAtomNumber()
+ // + " Last atom index " + curAtom.group.lastAtomIndex);
+ if (chainTerMap == null || prevAtom == null)
+ {
+ return true;
+ }
+ String curAtomChId = curAtom.getChainIDStr();
+ String prevAtomChId = prevAtom.getChainIDStr();
+ // new chain encoutered
+ if (!prevAtomChId.equals(curAtomChId))
+ {
+ // On chain switch add previous chain termination to xTerMap if not exists
+ if (!chainTerMap.containsKey(prevAtomChId))
+ {
+ chainTerMap.put(prevAtomChId, prevAtom);
+ }
+ // if current atom belongs to an already terminated chain and the resNum
+ // diff < 5 then mark as valid and update termination Atom
+ if (chainTerMap.containsKey(curAtomChId))
+ {
+ if (curAtom.getResno() < chainTerMap.get(curAtomChId).getResno())
+ {
+ return false;
+ }
+ if ((curAtom.getResno() - chainTerMap.get(curAtomChId).getResno()) < 5)
+ {
+ chainTerMap.put(curAtomChId, curAtom);
+ return true;
+ }
+ return false;
+ }
+ }
+ // atom with previously terminated chain encountered
+ else if (chainTerMap.containsKey(curAtomChId))
+ {
+ if (curAtom.getResno() < chainTerMap.get(curAtomChId).getResno())
+ {
+ return false;
+ }
+ if ((curAtom.getResno() - chainTerMap.get(curAtomChId).getResno()) < 5)
+ {
+ chainTerMap.put(curAtomChId, curAtom);
+ return true;
+ }
+ return false;
+ }
+ // HETATM with resNum jump > 2
+ return !(curAtom.isHetero() && ((curAtom.getResno() - prevAtom
+ .getResno()) > 2));
+ }
+
+ private void createAnnotation(SequenceI sequence, PDBChain chain,
+ org.jmol.modelset.Atom[] jmolAtoms)
+ {
+ char[] secstr = new char[sequence.getLength()];
+ char[] secstrcode = new char[sequence.getLength()];
+
+ // Ensure Residue size equals Seq size
+ if (chain.residues.size() != sequence.getLength())
+ {
+ return;
+ }
+ int annotIndex = 0;
+ for (Residue residue : chain.residues)
+ {
+ Atom repAtom = residue.getAtoms().get(0);
+ STR proteinStructureSubType = jmolAtoms[repAtom.atomIndex].group
+ .getProteinStructureSubType();
+ setSecondaryStructure(proteinStructureSubType, annotIndex, secstr,
+ secstrcode);
+ ++annotIndex;
+ }
+ addSecondaryStructureAnnotation(chain.pdbid, sequence, secstr,
+ secstrcode, chain.id, sequence.getStart());
+ }
+
+ /**
+ * 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')
+ {
+ try
+ {
+ asecstr[p] = new Annotation(String.valueOf(secstr[p]), null,
+ secstrcode[p], Float.NaN);
+ ssFound = true;
+ } catch (Exception e)
+ {
+ // e.printStackTrace();
+ }
+ }
+ }
+
+ 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);
+ }
+ }
+
private void waitForScript(Viewer jmd)
{
while (jmd.isScriptExecuting())
* Not implemented - returns null
*/
@Override
- public Dimension resizeInnerPanel(String data)
+ public int[] resizeInnerPanel(String data)
{
return null;
}
return null;
}
- /**
- * 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
- {
-
- chains = new ArrayList<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)
- {
- parseBiopolymers(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 parseBiopolymers(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;
- }
- }
- }
-
- /**
- * 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);
-
- addChainMetaData(sq, monomers, chainId);
-
- /*
- * add secondary structure predictions (if any)
- */
- if (isPredictSecondaryStructure())
- {
- addSecondaryStructureAnnotation(modelTitle, sq, secstr, secstrcode,
- chainId, firstResNum);
- }
-
- }
-
- public void addChainMetaData(SequenceI sq, List<Monomer> monomers,
- String chainId)
- {
- for (char res : sq.getSequence())
- {
-
- }
- }
-
- /**
- * 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)
- {
- List<Integer> chainLengths = new ArrayList<Integer>();
- int lastChainId = -1;
- int length = 0;
-
- for (Monomer monomer : monomers)
- {
- int chainId = monomer.chain.chainID;
- if (chainId != lastChainId && length > 0)
- {
- /*
- * change of chain - record the length of the last one
- */
- chainLengths.add(length);
- length = 0;
- }
- lastChainId = chainId;
- length++;
- }
- if (length > 0)
- {
- /*
- * record the length of the final chain
- */
- chainLengths.add(length);
- }
-
- return chainLengths;
- }
-
- /**
- * Returns a flattened list of Monomer (residues) in order, across all
- * BioPolymers in the model. This simplifies assembling chains which span
- * BioPolymers. The result omits any alternate residues reported for the same
- * sequence position (RESNUM value).
- *
- * @param ms
- * @param model
- * @return
- */
- protected List<Monomer> getMonomers(ModelSet ms, BioModel model)
- {
- List<Monomer> result = new ArrayList<Monomer>();
- int lastResNo = Integer.MIN_VALUE;
-
- for (BioPolymer bp : model.bioPolymers)
- {
- for (int groupLeadAtoms : bp.getLeadAtomIndices())
- {
- Group group = ms.at[groupLeadAtoms].group;
- if (group instanceof Monomer)
- {
- /*
- * ignore alternate residue at same position
- * example: 1ejg has residues A:LEU, B:ILE at RESNUM=25
- */
- int resNo = group.getResno();
- if (lastResNo != resNo)
- {
- result.add((Monomer) group);
- }
- lastResNo = resNo;
- }
- }
- }
- return result;
- }
-
public boolean isPredictSecondaryStructure()
{
return predictSecondaryStructure;
this.predictSecondaryStructure = predictSecondaryStructure;
}
- public Collection<PDBChain> getChains()
+ public boolean isVisibleChainAnnotation()
{
- return chains;
+ return visibleChainAnnotation;
}
- public void setChains(Collection<PDBChain> chains)
+ public void setVisibleChainAnnotation(boolean visibleChainAnnotation)
{
- this.chains = chains;
+ this.visibleChainAnnotation = visibleChainAnnotation;
}
}
git://source.jalview.org / jalview.git / blobdiff