use of de.bioforscher.jstructure.model.structure.aminoacid.AminoAcid in project jstructure by JonStargaryen.
the class Start2FoldXmlParser method parseSpecificExperiment.
public static void parseSpecificExperiment(Chain chain, InputStream inputStream, List<Integer> experimentIds) {
// experimentIds);
try {
// assign baseline resp. entry container for each residue
chain.aminoAcids().forEach(aminoAcid -> aminoAcid.getFeatureContainer().addFeature(new Start2FoldResidueAnnotation()));
Document document = Jsoup.parse(inputStream, "UTF-8", "/");
Elements experimentElements = document.getElementsByTag("experiment");
List<Experiment> experiments = experimentElements.stream().map(Experiment::parse).filter(experiment -> experimentIds.contains(experiment.getExperimentId())).collect(Collectors.toList());
if (experimentIds.size() != experiments.size()) {
throw new IllegalArgumentException("did not find all referenced experiments");
}
for (Experiment experiment : experiments) {
assignValuesForEarly(experiment, chain);
}
} catch (IOException e) {
throw new UncheckedIOException(e);
}
}
use of de.bioforscher.jstructure.model.structure.aminoacid.AminoAcid in project jstructure by JonStargaryen.
the class Start2FoldXmlParser method assignValuesForEarly.
private static void assignValuesForEarly(Experiment experiment, Chain chain) {
String pdbSequence = chain.getAminoAcidSequence();
String experimentSequence = experiment.getSequence();
// align sequences to ensure correct mapping
SequencePair<ProteinSequence, AminoAcidCompound> pair = null;
try {
pair = Alignments.getPairwiseAlignment(new ProteinSequence(experimentSequence), new ProteinSequence(pdbSequence), Alignments.PairwiseSequenceAlignerType.GLOBAL, new SimpleGapPenalty(), SubstitutionMatrixHelper.getBlosum62());
List<AminoAcid> aminoAcids = chain.aminoAcids().collect(Collectors.toList());
for (Experiment.Residue residue : experiment.getResidues()) {
int experimentIndex = residue.getIndex() - 1;
try {
int pdbIndex;
if (residue.getCode().equals("P") && residue.getIndex() == 1) {
// super-russian fix for STF0017 where the alignment should match theoretically
pdbIndex = 0;
} else {
pdbIndex = pair.getIndexInTargetForQueryAt(experimentIndex);
}
AminoAcid aminoAcid = aminoAcids.get(pdbIndex);
// assign experiment-specific protection level to residue
aminoAcid.getFeature(Start2FoldResidueAnnotation.class).addProtectionLevelEntry(ProtectionLevel.EARLY);
} catch (Exception e) {
// residue not present in structure - e.g. for STF0031 and STF0032
logger.warn("alignment:{}{}", System.lineSeparator(), pair.toString());
logger.warn("failed to map residue {}-{}", residue.getCode(), residue.getIndex(), e);
}
}
} catch (CompoundNotFoundException e) {
throw new IllegalArgumentException(e);
}
}
use of de.bioforscher.jstructure.model.structure.aminoacid.AminoAcid in project jstructure by JonStargaryen.
the class A01_CreateContactMaps method handleFile.
private static void handleFile(String line) {
String[] split = line.split(";");
String stfId = split[0];
String pdbId = split[1];
List<Integer> experimentIds = Pattern.compile(",").splitAsStream(split[2].replaceAll("\\[", "").replaceAll("]", "")).map(Integer::valueOf).collect(Collectors.toList());
Structure structure = StructureParser.fromPath(BASE_PATH.resolve("pdb").resolve(stfId + ".pdb")).parse();
Chain chain = structure.chainsWithAminoAcids().findFirst().get();
String sequence = chain.getAminoAcidSequence();
String secondaryStructureString = chain.aminoAcids().map(aminoAcid -> aminoAcid.getFeature(GenericSecondaryStructure.class)).map(GenericSecondaryStructure::getSecondaryStructure).map(SecondaryStructureType::getReducedRepresentation).collect(Collectors.joining()).toUpperCase();
Start2FoldConstants.write(BASE_PATH.resolve("reconstruction").resolve("fasta").resolve(stfId + ".fasta"), ">" + stfId + System.lineSeparator() + sequence);
Start2FoldConstants.write(BASE_PATH.resolve("reconstruction").resolve("sse").resolve(stfId + ".sse"), ">" + stfId + System.lineSeparator() + secondaryStructureString);
Start2FoldXmlParser.parseSpecificExperiment(chain, Start2FoldConstants.XML_DIRECTORY.resolve(stfId + ".xml"), experimentIds);
List<AminoAcid> aminoAcids = chain.aminoAcids().collect(Collectors.toList());
List<AminoAcid> earlyFoldingResidues = chain.aminoAcids().filter(aminoAcid -> aminoAcid.getFeature(Start2FoldResidueAnnotation.class).isEarly()).collect(Collectors.toList());
List<Pair<AminoAcid, AminoAcid>> contacts = SetOperations.unorderedPairsOf(aminoAcids).filter(pair -> areNonCovalentGroups(pair.getLeft(), pair.getRight())).filter(pair -> ProteinGraphFactory.InteractionScheme.CALPHA8.areInContact(pair.getLeft(), pair.getRight())).collect(Collectors.toList());
List<Pair<AminoAcid, AminoAcid>> earlyFoldingContacts = contacts.stream().filter(pair -> earlyFoldingResidues.contains(pair.getLeft()) && earlyFoldingResidues.contains(pair.getRight())).collect(Collectors.toList());
String percentage = StandardFormat.formatToInteger(100 * earlyFoldingContacts.size() / (double) contacts.size());
System.out.println("fraction of EFR contacts is " + percentage + "%: " + earlyFoldingContacts.size() + " " + contacts.size());
Start2FoldConstants.write(MAP_PATH.resolve(stfId + "-sampled-100-1.rr"), composeRRString(contacts, sequence));
Start2FoldConstants.write(MAP_PATH.resolve(stfId + "-efr-" + percentage + "-1.rr"), composeRRString(earlyFoldingContacts, sequence));
for (int i = 5; i < 100; i = i + 5) {
int numberOfContactsToSelect = (int) (i / (double) 100 * contacts.size());
for (int j = 1; j < 6; j++) {
Collections.shuffle(contacts);
List<Pair<AminoAcid, AminoAcid>> selectedContacts = contacts.subList(0, numberOfContactsToSelect);
Start2FoldConstants.write(MAP_PATH.resolve(stfId + "-random-" + i + "-" + j + ".rr"), composeRRString(selectedContacts, sequence));
}
}
// create samplings of random residues
for (int i = 5; i < 100; i = i + 5) {
int numberOfResiduesToSelect = (int) (i / (double) 100 * aminoAcids.size());
for (int j = 1; j < 6; j++) {
Collections.shuffle(aminoAcids);
List<AminoAcid> selectedAminoAcids = aminoAcids.subList(0, numberOfResiduesToSelect);
Start2FoldConstants.write(MAP_PATH.resolve(stfId + "-residues-" + i + "-" + j + ".rr"), composeRRString(contacts.stream().filter(contact -> selectedAminoAcids.contains(contact.getLeft()) && selectedAminoAcids.contains(contact.getRight())).collect(Collectors.toList()), sequence));
}
}
// create samplings of comparable nature of EFR contacts
for (int j = 1; j < 6; j++) {
int numberOfResiduesToSelect = earlyFoldingResidues.size();
List<AminoAcid> interactingResidues = getInteractingResidues(aminoAcids, contacts, numberOfResiduesToSelect);
Start2FoldConstants.write(MAP_PATH.resolve(stfId + "-interacting-" + percentage + "-" + j + ".rr"), composeRRString(contacts.stream().filter(contact -> interactingResidues.contains(contact.getLeft()) && interactingResidues.contains(contact.getRight())).collect(Collectors.toList()), sequence));
}
// create bin samplings of comparable nature of EFR contacts
for (int i = 5; i < 100; i = i + 5) {
int numberOfResiduesToSelect = (int) (i / (double) 100 * aminoAcids.size());
for (int j = 1; j < 6; j++) {
List<AminoAcid> interactingResidues = getInteractingResidues(aminoAcids, contacts, numberOfResiduesToSelect);
Start2FoldConstants.write(MAP_PATH.resolve(stfId + "-interacting2-" + percentage + "-" + j + ".rr"), composeRRString(contacts.stream().filter(contact -> interactingResidues.contains(contact.getLeft()) && interactingResidues.contains(contact.getRight())).collect(Collectors.toList()), sequence));
}
}
}
use of de.bioforscher.jstructure.model.structure.aminoacid.AminoAcid in project jstructure by JonStargaryen.
the class A02_CreatePyMolRenderJobsForStrongResidues method composePyMolCommand.
private static Optional<String> composePyMolCommand(Path path) {
try {
String entryId = path.toFile().getName().split("\\.")[0];
String pdbId = Jsoup.parse(path.toFile(), "UTF-8").getElementsByTag("protein").attr("pdb_id");
Structure structure = StructureParser.fromPdbId(pdbId).parse();
Chain chain = structure.chains().findFirst().get();
Start2FoldXmlParser.parse(chain, Start2FoldConstants.XML_DIRECTORY.resolve(entryId + ".xml"));
List<Integer> strongResidues = chain.aminoAcids().filter(aminoAcid -> aminoAcid.getFeature(Start2FoldResidueAnnotation.class).isStrong()).map(AminoAcid::getResidueIdentifier).map(ResidueIdentifier::getResidueNumber).collect(Collectors.toList());
if (strongResidues.isEmpty()) {
return Optional.empty();
}
return Optional.of("delete all" + System.lineSeparator() + "fetch " + pdbId + ", async=0" + System.lineSeparator() + // hide non-relevant stuff
"hide everything" + System.lineSeparator() + "show cartoon, chain A" + System.lineSeparator() + // decolor everything
"color grey80" + System.lineSeparator() + "zoom (chain A)" + System.lineSeparator() + strongResidues.stream().map(res -> "color efr, resi " + res).collect(Collectors.joining(System.lineSeparator())) + System.lineSeparator() + "ray" + System.lineSeparator() + "png " + Start2FoldConstants.PYMOL_DIRECTORY.resolve(entryId + "-strong.png") + System.lineSeparator());
} catch (IOException e) {
return Optional.empty();
}
}
use of de.bioforscher.jstructure.model.structure.aminoacid.AminoAcid in project jstructure by JonStargaryen.
the class A07_WriteStructuralInformationByContactCsv method handleLine.
private static Optional<String> handleLine(String line) {
try {
System.out.println(line);
String[] split = line.split(";");
String entryId = split[0];
String pdbId = split[1];
List<Integer> experimentIds = Pattern.compile(",").splitAsStream(split[2].replaceAll("\\[", "").replaceAll("]", "")).map(Integer::valueOf).collect(Collectors.toList());
// boolean sane = split[6].equalsIgnoreCase("true");
Structure structure = StructureParser.fromPdbId(pdbId).parse();
Chain chain = structure.chains().findFirst().get();
LinearAlgebra.PrimitiveDoubleArrayLinearAlgebra centroid = chain.calculate().centroid();
Path start2foldXml = Start2FoldConstants.XML_DIRECTORY.resolve(entryId + ".xml");
Start2FoldXmlParser.parseStability(chain, start2foldXml);
Start2FoldXmlParser.parseSpecificExperiment(chain, start2foldXml, experimentIds);
List<AminoAcid> earlyFoldingResidues = chain.aminoAcids().filter(aminoAcid -> aminoAcid.getFeature(Start2FoldResidueAnnotation.class).isEarly()).collect(Collectors.toList());
List<Integer> functionalResidueNumbers = Start2FoldConstants.extractFunctionalResidueNumbers(split);
List<AminoAcid> functionalResidues = new ArrayList<>();
// do nothing if no annotation of functional residues exists
if (!functionalResidueNumbers.isEmpty()) {
FunctionalResidueParser.parse(chain, functionalResidueNumbers);
chain.aminoAcids().filter(aminoAcid -> aminoAcid.getFeature(FunctionalResidueAnnotation.class).isFunctional()).forEach(functionalResidues::add);
}
List<AminoAcid> strongResidues = chain.aminoAcids().filter(aminoAcid -> aminoAcid.getFeature(Start2FoldResidueAnnotation.class).isStrong()).collect(Collectors.toList());
List<AminoAcid> orderedResidues = chain.aminoAcids().filter(aminoAcid -> !aminoAcid.getFeature(GenericSecondaryStructure.class).getSecondaryStructure().isCoilType()).collect(Collectors.toList());
List<AminoAcid> buriedResidues = chain.aminoAcids().filter(aminoAcid -> aminoAcid.getFeature(AccessibleSurfaceArea.class).isBuried()).collect(Collectors.toList());
List<AminoAcid> residuesInEarlyFoldingSecondaryStructureElements = chain.aminoAcids().filter(aminoAcid -> !aminoAcid.getFeature(GenericSecondaryStructure.class).getSecondaryStructure().isCoilType()).filter(aminoAcid -> {
GenericSecondaryStructure.SecondaryStructureElement surroundingSecondaryStructureElement = aminoAcid.getFeature(GenericSecondaryStructure.class).getSurroundingSecondaryStructureElement(aminoAcid);
List<AminoAcid> surroundingAminoAcids = chain.getAminoAcids().subList(surroundingSecondaryStructureElement.getStart(), surroundingSecondaryStructureElement.getEnd() + 1);
return surroundingAminoAcids.stream().anyMatch(earlyFoldingResidues::contains);
}).collect(Collectors.toList());
List<AminoAcid> aromaticResidues = chain.aminoAcids().filter(AminoAcid.Filter.AROMATIC).collect(Collectors.toList());
List<ContactStructuralInformation> contactStructuralInformation = StructuralInformationParserService.getInstance().parseContactStructuralInformation(Start2FoldConstants.DATA_DIRECTORY.resolve("si").resolve("raw").resolve(entryId.toUpperCase() + ".out"), chain, earlyFoldingResidues);
ResidueGraph conventionalProteinGraph = ResidueGraph.createResidueGraph(chain, ContactDefinitionFactory.createAlphaCarbonContactDefinition(8.0));
ResidueGraphCalculations residueGraphCalculations = new ResidueGraphCalculations(conventionalProteinGraph);
try {
EvolutionaryCouplingParser.parsePlmScore(contactStructuralInformation, Jsoup.parse(Start2FoldConstants.newInputStream(Start2FoldConstants.COUPLING_DIRECTORY.resolve(entryId + "_ec.html")), "UTF-8", ""), chain.getAminoAcids().size());
} catch (Exception e) {
}
boolean ecAnnotation = contactStructuralInformation.stream().anyMatch(csi -> csi.getPlmScore() != 0.0);
PLIPInteractionContainer plipInteractionContainer = chain.getFeature(PLIPInteractionContainer.class);
System.out.println("efr: " + (earlyFoldingResidues.size() > 0) + " strong: " + (strongResidues.size() > 0) + " functional: " + (functionalResidues.size() > 0) + " couplings: " + ecAnnotation);
return Optional.of(contactStructuralInformation.stream().map(contact -> {
AminoAcid aminoAcid1 = chain.select().residueNumber(contact.getResidueIdentifier1()).asAminoAcid();
AminoAcid aminoAcid2 = chain.select().residueNumber(contact.getResidueIdentifier2()).asAminoAcid();
Pair<AminoAcid, AminoAcid> pair = new Pair<>(aminoAcid1, aminoAcid2);
ResidueTopologicPropertiesContainer residueTopologicPropertiesContainer1 = aminoAcid1.getFeature(ResidueTopologicPropertiesContainer.class);
ResidueTopologicPropertiesContainer residueTopologicPropertiesContainer2 = aminoAcid1.getFeature(ResidueTopologicPropertiesContainer.class);
LinearAlgebra.PrimitiveDoubleArrayLinearAlgebra contactCentroid = aminoAcid1.calculate().centroid().add(aminoAcid2.calculate().centroid()).divide(2);
return pdbId + "," + "A" + "," + contact.getResidueIdentifier1() + "," + contact.getAa1() + "," + contact.getResidueIdentifier2() + "," + contact.getAa2() + "," + contact.getContactDistanceBin() + "," + (contact.getContactDistanceBin() == ContactDistanceBin.LONG) + "," + (contact.getContactDistanceBin() == ContactDistanceBin.MEDIUM) + "," + (contact.getContactDistanceBin() == ContactDistanceBin.SHORT) + "," + StandardFormat.format(contactCentroid.distance(centroid)) + "," + StandardFormat.format(contact.getAverageRmsdIncrease()) + "," + StandardFormat.format(contact.getAverageTmScoreIncrease()) + "," + StandardFormat.format(contact.getAverageQIncrease()) + "," + StandardFormat.format(contact.getMaximumRmsdIncrease()) + "," + StandardFormat.format(contact.getMaximumTmScoreIncrease()) + "," + StandardFormat.format(contact.getMaximumQIncrease()) + "," + StandardFormat.format(contact.getAverageRmsdIncreaseZScore()) + "," + contact.getFractionOfTopScoringContacts() + "," + StandardFormat.format(contact.getPlmScore()) + "," + contact.getCouplingRank() + "," + contact.istop02() + "," + contact.isTop04() + "," + contact.isTop06() + "," + contact.isTop08() + "," + contact.isTop10() + "," + contact.isTop12() + "," + contact.isTop14() + "," + contact.isTop16() + "," + StandardFormat.format(residueGraphCalculations.betweenness(pair)) + "," + StandardFormat.format(0.5 * residueTopologicPropertiesContainer1.getConventional().getBetweenness() + 0.5 * residueTopologicPropertiesContainer2.getConventional().getBetweenness()) + "," + StandardFormat.format(0.5 * residueTopologicPropertiesContainer1.getConventional().getCloseness() + 0.5 * residueTopologicPropertiesContainer2.getConventional().getCloseness()) + "," + StandardFormat.format(0.5 * residueTopologicPropertiesContainer1.getConventional().getClusteringCoefficient() + 0.5 * residueTopologicPropertiesContainer2.getConventional().getClusteringCoefficient()) + "," + plipInteractionContainer.getHydrogenBonds().stream().anyMatch(hydrogenBond -> isContact(hydrogenBond, aminoAcid1, aminoAcid2)) + "," + plipInteractionContainer.getHydrophobicInteractions().stream().anyMatch(hydrophobicInteraction -> isContact(hydrophobicInteraction, aminoAcid1, aminoAcid2)) + "," + contact.isEarlyFoldingResidue() + "," + contact.isEarlyFoldingContact() + "," + residueIsInCollection(functionalResidues, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + contactIsInCollection(functionalResidues, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + residueIsInCollection(strongResidues, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + contactIsInCollection(strongResidues, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + residueIsInCollection(buriedResidues, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + contactIsInCollection(buriedResidues, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + residueIsInCollection(orderedResidues, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + contactIsInCollection(orderedResidues, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + residueIsInCollection(residuesInEarlyFoldingSecondaryStructureElements, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + contactIsInCollection(residuesInEarlyFoldingSecondaryStructureElements, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + residueIsInCollection(aromaticResidues, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + contactIsInCollection(aromaticResidues, contact.getResidueIdentifier1(), contact.getResidueIdentifier2()) + "," + (earlyFoldingResidues.size() > 0) + "," + (strongResidues.size() > 0) + "," + (functionalResidues.size() > 0) + "," + ecAnnotation;
}).collect(Collectors.joining(System.lineSeparator())));
} catch (Exception e) {
logger.info("calculation failed for {}\nby: {}", line, e.getMessage());
return Optional.empty();
}
}
Aggregations