public static void main(String[] args) {
Point3d[] pts = SecondaryStructTools.obtain(NAME);
if (smooth) pts = CS.getSmoothedPoints(pts);
SecondaryStruct s = new SecondaryStruct(pts, smooth);
File f = new File("data/smooth_" + NAME + ".pdb");
Chain c = new ChainImpl();
c.setChainID(NAME.split("\\.")[1]);
try (PrintWriter pw = new PrintWriter(f)) {
for (int i = 0; i < pts.length; i++) {
Atom a = new AtomImpl();
a = new AtomImpl();
a.setName(CA_NAME);
a.setAltLoc(' ');
Group g = new AminoAcidImpl();
g.setPDBName(GROUP_NAME);
g.addAtom(a);
g.setResidueNumber(chainID, i + 1, null);
c.addGroup(g);
a.setX(pts[i].x);
a.setY(pts[i].y);
a.setZ(pts[i].z);
pw.print(a.toPDB());
}
} catch (FileNotFoundException e) {
e.printStackTrace();
}
// SecondaryStructureSequenceFeature sf = s.getSequenceFeature();
System.out.println("Start");
System.out.println(NAME);
System.out.println(s.getAlphaLength());
try (Scanner scan = new Scanner(System.in)) {
String in;
while (!(in = scan.next()).equals("X")) {
if (in.equals("g")) {
int st = scan.nextInt();
System.out.println(
SecondaryStructTools.distsToString(s.getRange(st - 1, scan.nextInt()), st));
} else if (in.equals("a")) s.printHelices();
else if (in.equals("b")) s.printStrands();
else if (in.equals("l")) System.out.println(s.length());
else if (in.equals("c")) s.printPoints();
// else if (in.equals("sf"))
// for (int i = 0; i < s.length(); i++)
// System.out.println((i + 1) + ":\t" + sf.toString(i));
else if (in.equals("test")) s.printVectors(s.getAlpha().getFeatures());
else if (in.equals("test1"))
System.out.println("=(0,0,0)\t=" + s.normP + "*50\t=" + s.normX + "*50");
else if (in.equals("test2"))
SecondaryStruct.printProjection(s.getAlphaNormProjection((byte) 0b00000000));
}
}
// sc.close();
}
/**
* Makes dummy CA atoms at 1A intervals. Only the x coordinate increments by one at each
* consecutive Atom.
*/
private Atom[] makeDummyCA(int len) {
Atom[] ca1;
Chain chain1 = new ChainImpl();
ca1 = new Atom[len];
for (int i = 0; i < len; i++) {
ca1[i] = new AtomImpl();
ca1[i].setName("CA");
ca1[i].setCoords(new double[] {i, 0, 0});
Group aa = new AminoAcidImpl();
aa.setPDBName("GLY");
aa.setResidueNumber(ResidueNumber.fromString(i + ""));
aa.addAtom(ca1[i]);
chain1.addGroup(aa);
}
return ca1;
}
private void recordUnidentifiableModifiedResidues(List<ModifiedCompound> modComps) {
Set<StructureGroup> identifiedComps = new HashSet<StructureGroup>();
for (ModifiedCompound mc : modComps) {
identifiedComps.addAll(mc.getGroups(true));
}
// TODO: use the ModifiedAminoAcid after Andreas add that.
for (Group group : residues) {
if (group.getType().equals(GroupType.HETATM)) {
StructureGroup strucGroup = StructureUtil.getStructureGroup(group, true);
if (!identifiedComps.contains(strucGroup)) {
unidentifiableModifiedResidues.add(strucGroup);
}
}
}
}
/**
* Record unidentifiable atom linkages in a chain. Only linkages between two residues or one
* residue and one ligand will be recorded.
*/
private void recordUnidentifiableAtomLinkages(
List<ModifiedCompound> modComps, List<Group> ligands) {
// first put identified linkages in a map for fast query
Set<StructureAtomLinkage> identifiedLinkages = new HashSet<StructureAtomLinkage>();
for (ModifiedCompound mc : modComps) {
identifiedLinkages.addAll(mc.getAtomLinkages());
}
// record
// cross link
int nRes = residues.size();
for (int i = 0; i < nRes - 1; i++) {
Group group1 = residues.get(i);
for (int j = i + 1; j < nRes; j++) {
Group group2 = residues.get(j);
List<Atom[]> linkages =
StructureUtil.findAtomLinkages(group1, group2, true, bondLengthTolerance);
for (Atom[] atoms : linkages) {
StructureAtomLinkage link =
StructureUtil.getStructureAtomLinkage(atoms[0], true, atoms[1], true);
unidentifiableAtomLinkages.add(link);
}
}
}
// attachment
int nLig = ligands.size();
for (int i = 0; i < nRes; i++) {
Group group1 = residues.get(i);
for (int j = 0; j < nLig; j++) {
Group group2 = ligands.get(j);
if (group1.equals(group2)) { // overlap between residues and ligands
continue;
}
List<Atom[]> linkages =
StructureUtil.findAtomLinkages(group1, group2, false, bondLengthTolerance);
for (Atom[] atoms : linkages) {
StructureAtomLinkage link =
StructureUtil.getStructureAtomLinkage(atoms[0], true, atoms[1], false);
unidentifiableAtomLinkages.add(link);
}
}
}
}
/**
* Returns the representation of a {@link ScopDomain} as a BioJava {@link Structure} object.
*
* @param domain a SCOP domain
* @param scopDatabase A {@link ScopDatabase} to use
* @param strictLigandHandling If set to false, hetero-atoms are included if and only if they
* belong to a chain to which the SCOP domain belongs; if set to true, hetero-atoms are
* included if and only if they are strictly within the definition (residue numbers) of the
* SCOP domain
* @return a Structure object
* @throws IOException
* @throws StructureException
*/
public Structure getStructureForDomain(
ScopDomain domain, ScopDatabase scopDatabase, boolean strictLigandHandling)
throws IOException, StructureException {
String pdbId = domain.getPdbId();
Structure fullStructure = getStructureForPdbId(pdbId);
Structure structure = domain.reduce(fullStructure);
// TODO It would be better to move all of this into the reduce method,
// but that would require ligand handling properties in StructureIdentifiers
// because ligands sometimes occur after TER records in PDB files, we may need to add some
// ligands back in
// specifically, we add a ligand if and only if it occurs within the domain
AtomPositionMap map = null;
List<ResidueRangeAndLength> rrs = null;
if (strictLigandHandling) {
map =
new AtomPositionMap(
StructureTools.getAllAtomArray(fullStructure), AtomPositionMap.ANYTHING_MATCHER);
rrs = ResidueRangeAndLength.parseMultiple(domain.getRanges(), map);
}
for (Chain chain : fullStructure.getChains()) {
if (!structure.hasChain(chain.getChainID())) {
continue; // we can't do anything with a chain our domain
}
// doesn't contain
Chain newChain = structure.getChainByPDB(chain.getChainID());
List<Group> ligands = StructureTools.filterLigands(chain.getAtomGroups());
for (Group group : ligands) {
boolean shouldContain = true;
if (strictLigandHandling) {
shouldContain = false; // whether the ligand occurs within the domain
for (ResidueRange rr : rrs) {
if (rr.contains(group.getResidueNumber(), map)) {
shouldContain = true;
}
}
}
boolean alreadyContains =
newChain.getAtomGroups().contains(group); // we don't want to add duplicate
// ligands
if (shouldContain && !alreadyContains) {
newChain.addGroup(group);
}
}
}
// build a more meaningful description for the new structure
StringBuilder header = new StringBuilder();
header.append(domain.getClassificationId());
if (scopDatabase != null) {
int sf = domain.getSuperfamilyId();
ScopDescription description = scopDatabase.getScopDescriptionBySunid(sf);
if (description != null) {
header.append(" | ");
header.append(description.getDescription());
}
}
structure.getPDBHeader().setDescription(header.toString());
return structure;
}
private static int append(SequenceFile.Writer writer, String pdbId, Structure s)
throws IOException {
int chainCount = 0;
for (Chain c : s.getChains()) {
List<Group> groups = c.getSeqResGroups(GroupType.AMINOACID);
System.out.println("seq len: " + c.getSeqResSequence().length());
System.out.println("seqresgroups.size(): " + c.getSeqResGroups(GroupType.AMINOACID).size());
List<Group> nAcids = c.getSeqResGroups(GroupType.NUCLEOTIDE);
boolean aminoAcid = true;
if (nAcids.size() > groups.size()) {
groups = nAcids;
aminoAcid = false;
}
int dna = 0;
int rna = 0;
int peptide = 0;
int dPeptide = 0;
int unknownResidues = 0;
Point3d[] coords = new Point3d[groups.size() * 3];
Integer[] sequence = new Integer[groups.size()];
int gaps = 0;
for (int i = 0; i < groups.size(); i++) {
Group g = groups.get(i);
char code = g.getChemComp().getOne_letter_code().charAt(0);
if (code == 'X') {
unknownResidues++;
}
sequence[i] = (int) code;
PolymerType p = g.getChemComp().getPolymerType();
if (p.equals(PolymerType.peptide)) {
peptide++;
} else if (p.equals(PolymerType.dpeptide)) {
dPeptide++;
} else if (p.equals(PolymerType.dna)) {
dna++;
} else if (p.equals(PolymerType.rna)) {
rna++;
}
Atom atom = null;
Atom atomN = null;
Atom atomC = null;
if (aminoAcid) {
atom = ((AminoAcidImpl) g).getCA();
atomN = ((AminoAcidImpl) g).getN();
atomC = ((AminoAcidImpl) g).getC();
} else {
atom = ((NucleotideImpl) g).getP();
}
if (atom == null || atomN == null || atomC == null) {
gaps++;
} else {
coords[i * 3] = new Point3d(atom.getCoords());
coords[i * 3 + 1] = new Point3d(atomN.getCoords());
coords[i * 3 + 2] = new Point3d(atomC.getCoords());
}
}
// ignore chains with less than 10 residues (with coordinates)
// System.out.println("size: " + (groups.size()-gaps));
if (groups.size() - gaps < 10) {
continue;
}
if (unknownResidues > (groups.size() - gaps) / 2) {
System.err.println("Polymer with many unknown residues ignored: " + pdbId + c.getChainID());
continue;
}
// ignore any mixed polymer types
if (dPeptide > 0) {
System.err.println("d-peptide ignored: " + pdbId + c.getChainID());
continue;
}
if (dna > 0 && rna > 0) {
System.err.println("DNA/RNA hybrid ignored: " + pdbId + c.getChainID());
continue;
}
// determine polymer type
SimplePolymerType polymerType = null;
if (peptide > 0) {
polymerType = SimplePolymerType.PROTEIN;
// System.out.println("PROTEIN: " + dna);
} else if (dna > 0) {
polymerType = SimplePolymerType.DNA;
// System.out.println("DNA: " + dna);
} else if (rna > 0) {
polymerType = SimplePolymerType.RNA;
// System.out.println("RNA: " + rna);
} else {
continue;
}
chainCount++;
Text key1 = new Text(pdbId + "." + c.getChainID());
ArrayWritable value1 = new IntArrayWritable();
value1.set(
SimplePolymerChainCodecHL.encodePolymerChain(
polymerType.ordinal(), coords, sequence, gaps));
writer.append(key1, value1);
}
return chainCount;
}
/** Get matched atoms for all linkages. */
private List<List<Atom[]>> getMatchedAtomsOfLinkages(
ModificationCondition condition, Map<Component, Set<Group>> mapCompGroups) {
List<ModificationLinkage> linkages = condition.getLinkages();
int nLink = linkages.size();
List<List<Atom[]>> matchedAtomsOfLinkages = new ArrayList<List<Atom[]>>(nLink);
for (int iLink = 0; iLink < nLink; iLink++) {
ModificationLinkage linkage = linkages.get(iLink);
Component comp1 = linkage.getComponent1();
Component comp2 = linkage.getComponent2();
// boolean isAA1 = comp1.;
// boolean isAA2 = comp2.getType()==true;
Set<Group> groups1 = mapCompGroups.get(comp1);
Set<Group> groups2 = mapCompGroups.get(comp2);
List<Atom[]> list = new ArrayList<Atom[]>();
List<String> potentialNamesOfAtomOnGroup1 = linkage.getPDBNameOfPotentialAtomsOnComponent1();
for (String name : potentialNamesOfAtomOnGroup1) {
if (name.equals("*")) {
// wildcard
potentialNamesOfAtomOnGroup1 = null; // search all atoms
break;
}
}
List<String> potentialNamesOfAtomOnGroup2 = linkage.getPDBNameOfPotentialAtomsOnComponent2();
for (String name : potentialNamesOfAtomOnGroup2) {
if (name.equals("*")) {
// wildcard
potentialNamesOfAtomOnGroup2 = null; // search all atoms
break;
}
}
for (Group g1 : groups1) {
for (Group g2 : groups2) {
if (g1.equals(g2)) {
continue;
}
// only for wildcard match of two residues
boolean ignoreNCLinkage =
potentialNamesOfAtomOnGroup1 == null
&& potentialNamesOfAtomOnGroup2 == null
&& residues.contains(g1)
&& residues.contains(g2);
Atom[] atoms =
StructureUtil.findNearestAtomLinkage(
g1,
g2,
potentialNamesOfAtomOnGroup1,
potentialNamesOfAtomOnGroup2,
ignoreNCLinkage,
bondLengthTolerance);
if (atoms != null) {
list.add(atoms);
}
}
}
if (list.isEmpty()) {
// broken linkage
break;
}
matchedAtomsOfLinkages.add(list);
}
return matchedAtomsOfLinkages;
}
/**
* @param modifications a set of {@link ProteinModification}s.
* @param residues
* @param ligands
* @param saveTo save result to
* @return map from component to list of corresponding residues in the chain.
*/
private void addModificationGroups(
final Set<ProteinModification> modifications,
final List<Group> residues,
final List<Group> ligands,
final Map<Component, Set<Group>> saveTo) {
if (residues == null || ligands == null || modifications == null) {
throw new IllegalArgumentException("Null argument(s).");
}
Map<Component, Set<Component>> mapSingleMultiComps = new HashMap<Component, Set<Component>>();
for (ProteinModification mod : modifications) {
ModificationCondition condition = mod.getCondition();
for (Component comp : condition.getComponents()) {
for (String pdbccId : comp.getPdbccIds()) {
Component single =
Component.of(Collections.singleton(pdbccId), comp.isNTerminal(), comp.isCTerminal());
Set<Component> mult = mapSingleMultiComps.get(single);
if (mult == null) {
mult = new HashSet<Component>();
mapSingleMultiComps.put(single, mult);
}
mult.add(comp);
}
}
}
{
// ligands
Set<Component> ligandsWildCard = mapSingleMultiComps.get(Component.of("*"));
for (Group group : ligands) {
String pdbccId = group.getPDBName().trim();
Set<Component> comps = mapSingleMultiComps.get(Component.of(pdbccId));
for (Component comp : unionComponentSet(ligandsWildCard, comps)) {
Set<Group> gs = saveTo.get(comp);
if (gs == null) {
gs = new LinkedHashSet<Group>();
saveTo.put(comp, gs);
}
gs.add(group);
}
}
}
{
// residues
if (residues.isEmpty()) {
return;
}
Set<Component> residuesWildCard = mapSingleMultiComps.get(Component.of("*"));
// for all residues
for (Group group : residues) {
String pdbccId = group.getPDBName().trim();
Set<Component> comps = mapSingleMultiComps.get(Component.of(pdbccId));
for (Component comp : unionComponentSet(residuesWildCard, comps)) {
Set<Group> gs = saveTo.get(comp);
if (gs == null) {
gs = new LinkedHashSet<Group>();
saveTo.put(comp, gs);
}
gs.add(group);
}
}
// for N-terminal
int nRes = residues.size();
int iRes = 0;
Group res;
do {
// for all ligands on N terminal and the first residue
res = residues.get(iRes++);
Set<Component> nTermWildCard = mapSingleMultiComps.get(Component.of("*", true, false));
Set<Component> comps = mapSingleMultiComps.get(Component.of(res.getPDBName(), true, false));
for (Component comp : unionComponentSet(nTermWildCard, comps)) {
Set<Group> gs = saveTo.get(comp);
if (gs == null) {
gs = new LinkedHashSet<Group>();
saveTo.put(comp, gs);
}
gs.add(res);
}
} while (iRes < nRes && ligands.contains(res));
// for C-terminal
iRes = residues.size() - 1;
do {
// for all ligands on C terminal and the last residue
res = residues.get(iRes--);
Set<Component> cTermWildCard = mapSingleMultiComps.get(Component.of("*", false, true));
Set<Component> comps = mapSingleMultiComps.get(Component.of(res.getPDBName(), false, true));
for (Component comp : unionComponentSet(cTermWildCard, comps)) {
Set<Group> gs = saveTo.get(comp);
if (gs == null) {
gs = new LinkedHashSet<Group>();
saveTo.put(comp, gs);
}
gs.add(res);
}
} while (iRes >= 0 && ligands.contains(res));
}
}