+ public static Regex getNewAlphafoldValidator()
+ {
+ Regex validator = new Regex("(AF-[A-Z]+[0-9]+[A-Z0-9]+-F1)");
+ validator.setIgnoreCase(true);
+ return validator;
+ }
+
+ PDBEntry.Type jmolFiletype = null;
+
+ /**
+ * resolve a jmol filetype string and update the jmolFiletype field
+ * accordingly
+ *
+ * @param jmolIdentifiedFileType
+ * @return true if filetype was identified as MMCIF, PDB
+ */
+ public boolean updateFileType(String jmolIdentifiedFileType)
+ {
+ if (jmolIdentifiedFileType == null
+ || jmolIdentifiedFileType.trim().equals(""))
+ {
+ return false;
+ }
+ if ("mmcif".equalsIgnoreCase(jmolIdentifiedFileType))
+ {
+ jmolFiletype = PDBEntry.Type.MMCIF;
+ return true;
+ }
+ if ("pdb".equalsIgnoreCase(jmolIdentifiedFileType))
+ {
+ jmolFiletype = PDBEntry.Type.PDB;
+ return true;
+ }
+ return false;
+ }
+
+ public void transformJmolModelToJalview(ModelSet ms) throws IOException
+ {
+ try
+ {
+ Regex alphaFold = getNewAlphafoldValidator();
+ String lastID = "";
+ List<SequenceI> rna = new ArrayList<SequenceI>();
+ List<SequenceI> prot = new ArrayList<SequenceI>();
+ PDBChain tmpchain;
+ String pdbId = (String) ms.getInfo(0, "title");
+ boolean isMMCIF = false;
+ String jmolFileType_String = (String) ms.getInfo(0, "fileType");
+ if (updateFileType(jmolFileType_String))
+ {
+ setStructureFileType(jmolFiletype.toString());
+ }
+
+ isMMCIF = PDBEntry.Type.MMCIF.equals(jmolFiletype);
+
+ if (pdbId == null)
+ {
+ setId(safeName(getDataName()));
+ setPDBIdAvailable(false);
+ }
+ else
+ {
+ setId(pdbId);
+ setPDBIdAvailable(true);
+ setAlphafoldModel(alphaFold.search(pdbId) && isMMCIF);
+ }
+ List<Atom> significantAtoms = convertSignificantAtoms(ms);
+ for (Atom tmpatom : significantAtoms)
+ {
+ if (tmpatom.resNumIns.trim().equals(lastID))
+ {
+ // phosphorylated protein - seen both CA and P..
+ continue;
+ }
+ tmpchain = findChain(tmpatom.chain);
+ if (tmpchain != null)
+ {
+ tmpchain.atoms.addElement(tmpatom);
+ }
+ else
+ {
+ AnnotationRowBuilder builder = null;
+ if (isAlphafoldModel()
+ || getTemperatureFactorType() == StructureImportSettings.TFType.PLDDT)
+ {
+ builder = new AlphaFoldAnnotationRowBuilder();
+ }
+
+ tmpchain = new PDBChain(getId(), tmpatom.chain, builder);
+ getChains().add(tmpchain);
+ tmpchain.atoms.addElement(tmpatom);
+ }
+ lastID = tmpatom.resNumIns.trim();
+ }
+ if (isParseImmediately())
+ {
+ // configure parsing settings from the static singleton
+ xferSettings();
+ }
+
+ makeResidueList();
+ makeCaBondList();
+
+ for (PDBChain chain : getChains())
+ {
+ SequenceI chainseq = postProcessChain(chain);
+ if (isRNA(chainseq))
+ {
+ rna.add(chainseq);
+ }
+ else
+ {
+ prot.add(chainseq);
+ }
+
+ // look at local setting for adding secondary tructure
+ if (predictSecondaryStructure)
+ {
+ createAnnotation(chainseq, chain, ms.at);
+ }
+ }
+ // if Alphafold, fetch the PAE matrix if doesn't already have one
+ if (isAlphafoldModel() && !hasPAEMatrix())
+ {
+ try
+ {
+ Console.info("retrieving pAE for " + pdbId);
+ File paeFile = EBIAlfaFold.fetchAlphaFoldPAE(pdbId, null);
+ this.setPAEMatrix(paeFile.getAbsolutePath());
+ } catch (Throwable t)
+ {
+ Console.error("Couldn't get the pAE for " + pdbId, t);
+ }
+ }
+ // add a PAEMatrix if set (either by above or otherwise)
+ if (hasPAEMatrix())
+ {
+ Alignment al = new Alignment(prot.toArray(new SequenceI[0]));
+ EBIAlfaFold.addAlphaFoldPAE(al, new File(this.getPAEMatrix()), 0,
+ null, false, false, null);
+
+ if (al.getAlignmentAnnotation() != null)
+ {
+ for (AlignmentAnnotation alann : al.getAlignmentAnnotation())
+ {
+ annotations.add(alann);
+ }
+ }
+ }
+ } 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 assignment 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)
+ {
+ int length = sq.getLength();
+ boolean ssFound = false;
+ Annotation asecstr[] = new Annotation[length + firstResNum - 1];
+ for (int p = 0; p < length; p++)
+ {
+ if (secstr[p] >= 'A' && secstr[p] <= 'z')
+ {
+ try
+ {
+ asecstr[p] = new Annotation(null, 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);
+ }
+ }
+