> hoseLookup;
- public boolean verbose = false;
- int maxSpheres = 6; //How many maximum spheres to use for the prediction
- boolean generatePictures = false;
- String picdir = null;
- String inFile = null;
- String hoseCodeFile = null;
- String hoseTSVfile = null;
-
- /**
- * Initializes a HOSECodePredictor by reading HOSECodes and assigned shift values from
- * from a tab-separated values (TSV) file, produced, for example, by the NMRShiftDBSDFParser
- * tool provided in this package.
- * Each line in the TSV file has the format
- *
- * String hosecode <\t>double shiftvalue
- *
- * @param hoseTSVfile tab-separated values (TSV) file with HOSECodes and assigned shift values
- * @throws Exception Exception if TSV file cannot be read
- */
- public HOSECodePredictor(String hoseTSVfile) throws Exception
- {
- readHOSECodeTable(hoseTSVfile);
- }
-
- /**
- * Initializes an empty HOSECodePredictor. Use readHOSECodeTable(String hoseTSVfile) to initialise.
- *
- * Each line in the TSV file has the format
- *
- * String hosecode <\t>double shiftvalue
- *
- * @param hoseTSVfile tab-separated values (TSV) file with HOSECodes and assigned shift values
- * @throws IOException Exception if TSV file cannot be read
- */
- public HOSECodePredictor() throws Exception
- {
-
- }
-
- public void predictFile(String outFile) throws Exception
- {
-
- IAtomContainer ac = null;
- File hoseTSVfile = new File(inFile);
- IteratingSDFReader iterator = new IteratingSDFReader(
- new FileReader(outFile),
- SilentChemObjectBuilder.getInstance()
- );
-
- while (iterator.hasNext())
- {
- ac = iterator.next();
- predict(ac);
- generatePicture(ac, picdir);
- }
- iterator.close();
- }
-
- public void readHOSECodeTable() throws Exception
- {
- readHOSECodeTable(this.hoseTSVfile);
- }
-
- /**
- * Reads HOSE code table from TSV file. Without this, no prediction can be performed.
- * Each line in the TSV file has the format
- *
- * String hosecode <\t>double shiftvalue
- *
- *
- * @param hoseTSVfile
- * @throws Exception
- */
- public void readHOSECodeTable(String hoseTSVfile) throws Exception
- {
- String line = null;
- StringTokenizer strtok;
- String hose;
- Double shift;
- ArrayList shifts;
- int linecounter = 0;
-
- if (verbose) System.out.println("Start reading HOSE codes from " + hoseTSVfile);
-
- BufferedReader br = new BufferedReader(new FileReader(hoseTSVfile));
- hoseLookup = new Hashtable>();
- while((line = br.readLine()) != null)
- {
- strtok = new StringTokenizer(line, "\t");
- linecounter++;
- hose = strtok.nextToken();
- shift = Double.parseDouble(strtok.nextToken());
- //System.out.println(hose + " ---- " + shift);
- if (hoseLookup.containsKey(hose))
- {
- shifts = hoseLookup.get(hose);
- shifts.add(new Double(shift));
- //System.out.println("HOSE code already in hashtable. Adding to existing list.");
- }
- else
- {
- //System.out.println("HOSE code not in hashtable. Adding to new list.");
- shifts = new ArrayList();
- shifts.add(new Double(shift));
- hoseLookup.put(hose, shifts);
- }
- }
- br.close();
- if (verbose) System.out.println("Finished reading " + linecounter + " lines of HOSE codes.");
-
- }
-
- /**
- * Predicts NMR chemical shifts based on a given HOSE code table read by the
- * Constructor of this class.
- * The predicted chemical shifts are assigned to each atom by
- * as a property of type CDKConstants.NMRSHIFT_CARBON.
- * They are also written as CDKConstants.COMMENT to aid the
- * DepictionGenerator class to generate a picture with shift annotations
- * for all carbon atoms.
- *
- * @param ac The IAtomContainer for which to predict the shift values
- * @throws Exception Thrown if something goes wrong.
- */
- public void predict(IAtomContainer ac) throws Exception
- {
- IAtom atom;
- String hose = null;
- Double shift = null;
- HOSECodeGenerator hcg = new HOSECodeGenerator();
- DecimalFormat df = new DecimalFormat();
- df.setMaximumFractionDigits(2);
- /**
- * A visual appendix to the CDKConstants.NMRSHIFT_COMMENT annotation to show
- * how many HOSE code spheres where used to predict this shift value.
- */
- String[] sphereCount =
- {
- "'",
- "''",
- "'''",
- "''''",
- "'''''",
- "'''''",
- "''''''",
- "'''''''"
- };
- fixExplicitHydrogens(ac);
- if (verbose) System.out.println("Entering prediction module");
- for (int f = 0; f < ac.getAtomCount(); f++)
- {
- atom = ac.getAtom(f);
- if (verbose) System.out.println("Atom no. " + f);
- if (atom.getAtomicNumber() == 6)
- {
- // We descend from N-sphere HOSE codes defined by maxSpheres to those with lower spheres
- for (int g = maxSpheres; g > 0; g--)
- {
- hose = hcg.getHOSECode(ac, atom, g);
- //System.out.println("Look-up for HOSE code " + hose);
- try{
- shift = getShift(hose);
- if (shift != null)
- {
- if (verbose) System.out.println("Shift " + df.format(shift) + " found with " + g + "-sphere HOSE code.");
- ac.getAtom(f).setProperty(CDKConstants.NMRSHIFT_CARBON, df.format(shift));
- ac.getAtom(f).setProperty(CDKConstants.COMMENT, df.format(shift) + sphereCount[g - 1]);
- //If we found a HOSE code of a higher sphere, we take that one and skip the lower ones
- break;
- }
- }
- catch(Exception e)
- {
- e.printStackTrace();
-
- }
- }
- }
- }
- }
-
- public Double getShift(String hose)
- {
- double shiftvalue = 0.0;
- if (!hoseLookup.containsKey(hose)) return null;
- ArrayList list = hoseLookup.get(hose);
- for (int f = 0; f < list.size(); f++)
- {
- shiftvalue = shiftvalue + ((Double) list.get(f)).doubleValue();
- }
- shiftvalue = shiftvalue / list.size();
- if (verbose) System.out.println("Predicted HOSE code from " + list.size() + " values");
-
- return new Double(shiftvalue);
- }
-
- public void generatePicture(IAtomContainer ac, String path) throws IOException, CDKException
- {
- String moleculeTitle = "";
- /* Path separators differ in operating systems. Unix uses slash, windows backslash.
- That's why Java offers this constant File.pathSeparator since it knows what OS it is running on */
- if (path.endsWith(File.separator))
- moleculeTitle = path + "mol.png";
- else
- moleculeTitle = path + File.separator + "mol.png";
- DepictionGenerator dg = new DepictionGenerator().withSize(800, 800).withAtomColors().withAtomValues().withMolTitle().withFillToFit();
- dg.depict(ac).writeTo(moleculeTitle);
- }
-
-
-
- /**
- * This predictor cannot handle explicit hydrogens. Where therefore convert them to implicit first
- */
- void fixExplicitHydrogens(IAtomContainer ac)
- {
- IAtom atomB;
- for (IAtom atomA : ac.atoms())
- {
- if (atomA.getAtomicNumber() == 1)
- {
- atomB = ac.getConnectedAtomsList(atomA).get(0);
- atomB.setImplicitHydrogenCount(atomB.getImplicitHydrogenCount() +1 );
- ac.removeAtom(atomA);
- }
- }
- }
-
- private void parseArgs(String[] args) throws ParseException
- {
- Options options = setupOptions(args);
- CommandLineParser parser = new DefaultParser();
- try {
- CommandLine cmd = parser.parse( options, args);
- this.inFile = cmd.getOptionValue("infile");
- this.hoseTSVfile = cmd.getOptionValue("hosecodes");
- if (cmd.hasOption("maxspheres"))
- {
- this.maxSpheres = Integer.parseInt(cmd.getOptionValue("maxspheres"));
- }
- if (cmd.hasOption("verbose")) this.verbose = true;
-
- if (cmd.hasOption("picdir"))
- {
- this.generatePictures = true;
- this.picdir = cmd.getOptionValue("picdir");
-
- }
- } catch (ParseException e) {
- // TODO Auto-generated catch block
- HelpFormatter formatter = new HelpFormatter();
- formatter.setOptionComparator(null);
- String header = "Predict NMR chemical shifts for a given molecule based on table of HOSE codes and assigned shifts.\n\n";
- String footer = "\nPlease report issues at https://github.com/steinbeck/spectra";
- formatter.printHelp( "java -jar casekit.jar casekit.HOSECodePredictor", header, options, footer, true );
- throw new ParseException("Problem parsing command line");
- }
- }
-
- private Options setupOptions(String[] args)
- {
- Options options = new Options();
- Option hosefile = Option.builder("s")
- .required(true)
- .hasArg()
- .longOpt("hosecodes")
- .desc("filename of TSV file with HOSE codes (required)")
- .build();
- options.addOption(hosefile);
- Option infile = Option.builder("i")
- .required(true)
- .hasArg()
- .longOpt("infile")
- .desc("filename of with SDF/MOL file of structures to be predicted (required)")
- .build();
- options.addOption(infile);
- Option verbose = Option.builder("v")
- .required(false)
- .longOpt("verbose")
- .desc("generate messages about progress of operation")
- .build();
- options.addOption(verbose);
- Option picdir = Option.builder("d")
- .required(false)
- .hasArg()
- .longOpt("picdir")
- .desc("store pictures of structures with assigned shifts in given directory")
- .build();
- options.addOption(picdir);
- Option maxspheres = Option.builder("m")
- .required(false)
- .hasArg()
- .longOpt("maxspheres")
- .desc("maximum sphere size up to which to generate HOSE codes. Default is 6 spheres if this option is ommitted.")
- .build();
- options.addOption(maxspheres);
- return options;
- }
-
- public static void main(String[] args) {
- // TODO Auto-generated method stub
- HOSECodePredictor hcp = null;
- try {
- hcp = new HOSECodePredictor();
- hcp.parseArgs(args);
- hcp.readHOSECodeTable();
- hcp.predictFile(hcp.inFile);
- } catch (Exception e) {
- // We don't do anything here. Apache CLI will print a usage text.
- if (hcp.verbose) e.printStackTrace();
- }
-
- }
-}
diff --git a/src/casekit/NMRShiftDBSDFParser.java b/src/casekit/NMRShiftDBSDFParser.java
deleted file mode 100644
index 796fe4c..0000000
--- a/src/casekit/NMRShiftDBSDFParser.java
+++ /dev/null
@@ -1,256 +0,0 @@
-/*
-* This Open Source Software is provided to you under the MIT License
- * Refer to doc/mit.license or https://opensource.org/licenses/MIT for more information
- *
- * Copyright (c) 2017, Christoph Steinbeck
- */
-package casekit;
-
-import java.io.BufferedWriter;
-import java.io.File;
-import java.io.FileOutputStream;
-import java.io.FileReader;
-import java.io.IOException;
-import java.io.OutputStreamWriter;
-import java.util.StringTokenizer;
-
-import org.apache.commons.cli.CommandLine;
-import org.apache.commons.cli.CommandLineParser;
-import org.apache.commons.cli.DefaultParser;
-import org.apache.commons.cli.HelpFormatter;
-import org.apache.commons.cli.Option;
-import org.apache.commons.cli.Options;
-import org.apache.commons.cli.ParseException;
-import org.openscience.cdk.CDKConstants;
-import org.openscience.cdk.aromaticity.Aromaticity;
-import org.openscience.cdk.depict.DepictionGenerator;
-import org.openscience.cdk.exception.CDKException;
-import org.openscience.cdk.interfaces.IAtomContainer;
-import org.openscience.cdk.interfaces.IMolecularFormula;
-import org.openscience.cdk.io.iterator.IteratingSDFReader;
-import org.openscience.cdk.silent.SilentChemObjectBuilder;
-import org.openscience.cdk.tools.HOSECodeGenerator;
-import org.openscience.cdk.tools.manipulator.AtomContainerManipulator;
-
-/**
- * Helper class to parse an NMRShiftDB SDF file with spectra assignments
- * and convert it to a tab-separated values file with HOSE codes
- * (Bremser, W., HOSE - A Novel Substructure Code, Analytica Chimica Acta, 1978, 103:355-365)
- * and associated shift values.
- * The TSV file can then be used by HOSECodePredictor to predict spectra
- *
- * @author Christoph Steinbeck
- */
-
-public class NMRShiftDBSDFParser {
- BufferedWriter bw;
- IMolecularFormula formula = null;
- String comment = null;
- String carbonNMR = null;
- String hydrogenNMR = null;
- String moleculeTitle = null;
- int carbonNMRCount = 0;
- int hydrogenNMRCount = 0;
- int moleculeCount = 0;
- String report = "";
- String temp = "";
- boolean generatePictures = false;
- String picdir = null;
- int hoseCodeCounter = 0;
- int carbonCounter = 0;
- String inFile = null;
- String outFile = null;
- boolean verbose = false;
- int maxSpheres;
-
- public NMRShiftDBSDFParser(String[] args) throws Exception
- {
- parseArgs(args);
- if (verbose) System.out.println("Starting HOSE code generation with " + maxSpheres + " sphere from " + inFile);
- File fout = new File(outFile);
- FileOutputStream fos = new FileOutputStream(fout);
- bw = new BufferedWriter(new OutputStreamWriter(fos));
-
- IAtomContainer ac = SilentChemObjectBuilder.getInstance().newAtomContainer();
-
- IteratingSDFReader iterator = new IteratingSDFReader(
- new FileReader(inFile),
- SilentChemObjectBuilder.getInstance()
- );
-
- while (iterator.hasNext())
- {
- ac = iterator.next();
- carbonNMR = (String)ac.getProperty("Spectrum 13C 0");
- hydrogenNMR = (String)ac.getProperty("Spectrum 1H 0");
- if (carbonNMR != null)
- {
- carbonNMRCount++;
- ac = assignCarbonNMR(ac, carbonNMR);
- generateHOSECodes(ac, maxSpheres);
- }
- if (hydrogenNMR != null) hydrogenNMRCount++;
- moleculeCount ++;
- if (generatePictures) generatePicture(ac, picdir);
- }
- iterator.close();
- report = "File contains " + moleculeCount + " molecules with " + carbonNMRCount + " carbon spectra and " + hydrogenNMRCount + " hydrogen spectra.\n";
- report += hoseCodeCounter + " HOSE codes generated for " + carbonCounter + "carbon atoms, and written to file.";
- if (verbose) System.out.println(report);
- bw.close();
- }
-
- IAtomContainer assignCarbonNMR(IAtomContainer ac, String nmrString) throws IOException, CDKException
- {
- String sigString = null, shiftString = null, multString = null, atomNumString = null;
- StringTokenizer strTok1 = new StringTokenizer(nmrString, "|");
- StringTokenizer strTok2 = null;
- while (strTok1.hasMoreTokens())
- {
- sigString = strTok1.nextToken(); //System.out.println(sigString);
- strTok2 = new StringTokenizer(sigString, ";");
- while (strTok2.hasMoreTokens())
- {
- shiftString = strTok2.nextToken(); //System.out.println(shiftString);
- multString = strTok2.nextToken(); //System.out.println(multString);
- atomNumString = strTok2.nextToken(); //System.out.println(atomNumString);
- try
- {
- ac.getAtom(Integer.parseInt(atomNumString)).setProperty(CDKConstants.NMRSHIFT_CARBON, Double.parseDouble(shiftString));
- ac.getAtom(Integer.parseInt(atomNumString)).setProperty(CDKConstants.COMMENT, Double.parseDouble(shiftString));
- }catch(Exception exc)
- {
- System.out.println("Failed to assign shift to atom no. " + Integer.parseInt(atomNumString) + " in molecule no " + moleculeCount + ", title: " + ac.getProperty(CDKConstants.TITLE) + " with " + ac.getAtomCount() + " atoms. ");
- }
- }
- }
- return ac;
- }
-
- public void generateHOSECodes(IAtomContainer ac, int maxSpheres) throws Exception
- {
- String hose = null;
- HOSECodeGenerator hcg = new HOSECodeGenerator();
- AtomContainerManipulator.percieveAtomTypesAndConfigureAtoms(ac);
- Aromaticity.cdkLegacy().apply(ac);
- for (int f = 0; f < ac.getAtomCount(); f++)
- {
- if (ac.getAtom(f).getAtomicNumber() == 6)
- {
- carbonCounter ++;
- for (int g = 0; g < maxSpheres; g++)
- {
- hose = hcg.getHOSECode(ac, ac.getAtom(f), g + 1);
- if (hose != null && ac.getAtom(f).getProperty(CDKConstants.NMRSHIFT_CARBON) != null)
- {
- bw.write(hose + "\t" + ac.getAtom(f).getProperty(CDKConstants.NMRSHIFT_CARBON));
- bw.newLine();
- hoseCodeCounter++;
- }
- }
- }
- }
- }
-
- public void generatePicture(IAtomContainer ac, String picdir) throws IOException, CDKException
- {
- try
- {
- temp = ac.getProperty(CDKConstants.TITLE);
- if (temp != null && temp.length() > 15) temp = temp.substring(0, 14) + "...";
- ac.setProperty(CDKConstants.TITLE, temp);
- }
- catch(Exception e)
- {
- System.out.println("Problem with title " + temp);
- }
- if (!picdir.endsWith(File.separator)) picdir += File.separator;
- moleculeTitle = picdir + String.format("%03d", moleculeCount) + "-mol.png";
- if (verbose) System.out.println(moleculeTitle);
- DepictionGenerator dg = new DepictionGenerator().withSize(800, 800).withAtomColors().withAtomValues().withMolTitle().withFillToFit();
- dg.depict(ac).writeTo(moleculeTitle);
- }
-
- private void parseArgs(String[] args) throws ParseException
- {
- Options options = setupOptions(args);
- CommandLineParser parser = new DefaultParser();
- try {
- CommandLine cmd = parser.parse( options, args);
- this.inFile = cmd.getOptionValue("infile");
- this.outFile = cmd.getOptionValue("outfile");
- if (cmd.hasOption("maxspheres"))
- {
- this.maxSpheres = Integer.parseInt(cmd.getOptionValue("maxspheres"));
- }
- if (cmd.hasOption("verbose")) this.verbose = true;
-
- if (cmd.hasOption("picdir"))
- {
- this.generatePictures = true;
- this.picdir = cmd.getOptionValue("picdir");
-
- }
- } catch (ParseException e) {
- // TODO Auto-generated catch block
- HelpFormatter formatter = new HelpFormatter();
-
- formatter.setOptionComparator(null);
- String header = "Generates a table of HOSE codes and assigned shifts from an NMRShiftDB SDF file from http://nmrshiftdb.nmr.uni-koeln.de/portal/js_pane/P-Help.\n\n";
- String footer = "\nPlease report issues at https://github.com/steinbeck/spectra";
- formatter.printHelp( "java -jar casekit.jar casekit.NMRShiftDBSDFParser", header, options, footer, true );
- throw new ParseException("Problem parsing command line");
- }
- }
-
- private Options setupOptions(String[] args)
- {
- Options options = new Options();
- Option infile = Option.builder("i")
- .required(true)
- .hasArg()
- .longOpt("infile")
- .desc("filename of NMRShiftDB SDF with spectra (required)")
- .build();
- options.addOption(infile);
- Option outfile = Option.builder("o")
- .required(true)
- .hasArg()
- .longOpt("outfile")
- .desc("filename of generated HOSE code table (required)")
- .build();
- options.addOption(outfile);
- Option verbose = Option.builder("v")
- .required(false)
- .longOpt("verbose")
- .desc("generate messages about progress of operation")
- .build();
- options.addOption(verbose);
- Option picdir = Option.builder("d")
- .required(false)
- .hasArg()
- .longOpt("picdir")
- .desc("store pictures in given directory")
- .build();
- options.addOption(picdir);
- Option maxspheres = Option.builder("m")
- .required(false)
- .hasArg()
- .longOpt("maxspheres")
- .desc("maximum sphere size up to which to generate HOSE codes")
- .build();
- options.addOption(maxspheres);
- return options;
- }
-
- public static void main(String[] args)
- {
- try {
- new NMRShiftDBSDFParser(args);
- } catch (Exception e) {
- // TODO Auto-generated catch block
- e.printStackTrace();
- }
- }
-
-}
diff --git a/src/casekit/Result.java b/src/casekit/Result.java
deleted file mode 100644
index e451948..0000000
--- a/src/casekit/Result.java
+++ /dev/null
@@ -1,37 +0,0 @@
-/*
-* This Open Source Software is provided to you under the MIT License
- * Refer to doc/mit.license or https://opensource.org/licenses/MIT for more information
- *
- * Copyright (c) 2017, Christoph Steinbeck
- */
-
-package casekit;
-
-import org.openscience.cdk.interfaces.IAtomContainer;
-
-public class Result {
- public IAtomContainer ac;
- public double score;
-
- public Result(IAtomContainer ac, double score) {
- super();
- this.ac = ac;
- this.score = score;
- }
-
- public IAtomContainer getAc() {
- return ac;
- }
- public void setAc(IAtomContainer ac) {
- this.ac = ac;
- }
- public double getScore() {
- return score;
- }
-
- public void setScore(double score) {
- this.score = score;
- }
-
-
-}
\ No newline at end of file
diff --git a/src/casekit/Signal.java b/src/casekit/Signal.java
deleted file mode 100644
index 1e87f1f..0000000
--- a/src/casekit/Signal.java
+++ /dev/null
@@ -1,45 +0,0 @@
-/*
-* This Open Source Software is provided to you under the MIT License
- * Refer to doc/mit.license or https://opensource.org/licenses/MIT for more information
- *
- * Copyright (c) 2017, Christoph Steinbeck
- */
-
-package casekit;
-
-public class Signal {
-
- Double shift = null;
- Integer mult = null;
-
- public Signal() {
- // TODO Auto-generated constructor stub
- }
-
- Signal(double shift)
- {
- setShift(shift);
- }
-
- Signal(double shift, int mult)
- {
- setShift(shift);
- setMult(mult);
- }
-
- public Integer getMult() {
- return mult;
- }
- public void setMult(Integer mult) {
- this.mult = mult;
- }
-
- public Double getShift() {
- return shift;
- }
-
- public void setShift(Double shift) {
- this.shift = shift;
- }
-
-}
diff --git a/src/casekit/SimilarityRanker.java b/src/casekit/SimilarityRanker.java
deleted file mode 100644
index a85d1b9..0000000
--- a/src/casekit/SimilarityRanker.java
+++ /dev/null
@@ -1,305 +0,0 @@
-
-/*
-* This Open Source Software is provided to you under the MIT License
- * Refer to doc/mit.license or https://opensource.org/licenses/MIT for more information
- *
- * Copyright (c) 2017, Christoph Steinbeck
- */
-package casekit;
-
-import java.io.BufferedReader;
-import java.io.File;
-import java.io.FileReader;
-import java.io.IOException;
-import java.text.DecimalFormat;
-import java.util.ArrayList;
-import java.util.Comparator;
-import java.util.StringTokenizer;
-
-import org.apache.commons.cli.CommandLine;
-import org.apache.commons.cli.CommandLineParser;
-import org.apache.commons.cli.DefaultParser;
-import org.apache.commons.cli.HelpFormatter;
-import org.apache.commons.cli.Option;
-import org.apache.commons.cli.Options;
-import org.apache.commons.cli.ParseException;
-import org.openscience.cdk.CDKConstants;
-import org.openscience.cdk.depict.DepictionGenerator;
-import org.openscience.cdk.interfaces.IAtom;
-import org.openscience.cdk.interfaces.IAtomContainer;
-import org.openscience.cdk.io.iterator.IteratingSDFReader;
-import org.openscience.cdk.silent.SilentChemObjectBuilder;
-
-/**
- * SimilarityRanker uses a SpectrumPredictor and parses an SDF file, returning a configurable number of compounds and
- * their ranked spectrum similarity.
- *
- * This Open Source Software is provided to you under the MIT License
- * Refer to doc/mit.license or https://opensource.org/licenses/MIT for more information
- *
- * Copyright (c) 2017, Christoph Steinbeck
- *
- * @author steinbeck
- *
- */
-public class SimilarityRanker {
-
- public boolean verbose = true;
- DecimalFormat df;
- public int resultListSize = 100;
- public String inFile = null;
- public String outPath = null;
- public String spectrumFile = null;
- public String hoseTSVFile = null;
- ArrayList spectrum = null;
- ArrayList results = null;
-
- public boolean isVerbose() {
- return verbose;
- }
-
- public void setVerbose(boolean verbose) {
- this.verbose = verbose;
- }
-
- public int getResultListSize() {
- return resultListSize;
- }
-
- public void setResultListSize(int resultListSize) {
- this.resultListSize = resultListSize;
- }
-
- public SimilarityRanker() {
- // TODO Auto-generated constructor stub
- df = new DecimalFormat();
- df.setMaximumFractionDigits(2);
- }
-
- public void readSpectrum() throws NumberFormatException, IOException
- {
- String line;
- StringTokenizer strtok;
- int linecounter = 0;
- Double shift = null;
- Integer mult = null;
- Signal signal;
- String tempString;
- ArrayList spectrum = new ArrayList();
- BufferedReader br = new BufferedReader(new FileReader(spectrumFile));
- if (verbose) System.out.println("Start reading spectrum from " + spectrumFile);
- while((line = br.readLine()) != null)
- {
- if (!line.startsWith("#") && line.trim().length() > 0)
- {
- strtok = new StringTokenizer(line, ";");
- if (verbose) System.out.println(line);
- linecounter++;
-
- shift = Double.parseDouble(strtok.nextToken().trim());
- mult = Integer.parseInt(strtok.nextToken().trim());
- signal = new Signal(shift, mult);
- spectrum.add(signal);
- }
- }
- br.close();
- if (verbose) System.out.println("Read " + linecounter + " signals from spectrum in file " + spectrumFile);
-
- this.spectrum = spectrum;
- }
-
-
- public ArrayList rank() throws Exception
- {
- /*
- * Iterate of SDF file given by input file, predict a spectrum and calculate a similarity with the
- * spectrum given in @spectrum.
- * Store the 10 most similar spectra in a list and write them to outFile in the end
- */
-
- HOSECodePredictor predictor = new HOSECodePredictor(hoseTSVFile);
- IAtomContainer ac = null;
- double similarity = 0.0;
- double bestSimilarity = 1000000000.0;
- results = new ArrayList();
- ResultComparator comp = new ResultComparator();
- IteratingSDFReader iterator = new IteratingSDFReader(
- new FileReader(inFile),
- SilentChemObjectBuilder.getInstance()
- );
-
- while (iterator.hasNext())
- {
- ac = iterator.next();
- predictor.predict(ac);
- similarity = calculateSimilarity(ac, spectrum);
- if (results.size() > 0)
- {
- if (similarity < results.get(results.size()-1).getScore())
- {
- bestSimilarity = similarity;
- ac.setProperty(CDKConstants.TITLE, "Distance " + df.format(similarity));
- results.add(new Result(ac, similarity));
- results.sort(comp);
- //After sorting, we remove the worst entry and thereby trim the results list to resultListSize
- if (results.size() == resultListSize) results.remove(resultListSize - 1);
- }
- }
- else results.add(new Result(ac, similarity));
- }
- iterator.close();
- if (verbose) System.out.println("Calculation finished. Best similarity = " + bestSimilarity);
- return results;
- }
-
- public double calculateSimilarity(IAtomContainer ac, ArrayList spectrum)
- {
- double similarity = 0.0;
- double lastDiff = 0.0;
- int counter = 0;
- String shift = null;
- boolean matchFound = false;
- double diff = 0.0;
- double shifts[] = new double[spectrum.size()];
- for (IAtom atom : ac.atoms())
- {
- if (atom.getAtomicNumber() == 6)
- {
- shift = atom.getProperty(CDKConstants.NMRSHIFT_CARBON);
- if (shift != null) shifts[counter] = Double.parseDouble(shift);
- else shifts[counter] = -1.0;
- counter ++;
- }
- }
- for (int f = 0; f < spectrum.size(); f++)
- {
- lastDiff = 10000000000.0;
- matchFound = false;
- for (int g = 0; g < spectrum.size(); g++)
- {
- if (shifts[f] > spectrum.get(g).getShift().doubleValue()) diff = shifts[f] - spectrum.get(g).getShift().doubleValue();
- else diff = spectrum.get(g).getShift().doubleValue() - shifts[f];
- df.format(diff);
- if (diff < lastDiff)
- {
- lastDiff = diff;
- matchFound = true;
- }
- }
- if (matchFound) similarity += lastDiff;
- }
- return similarity/spectrum.size();
- }
-
- public void reportResults() throws Exception
- {
- String filename = null;
- DepictionGenerator dg = null;
- if (!outPath.endsWith(File.separator))
- outPath += File.separator;
- for (int f = 0; f < results.size(); f++)
- {
- filename = outPath + String.format("%03d", f) + "-mol.png";
- dg = new DepictionGenerator().withSize(800, 800).withAtomColors().withAtomValues().withMolTitle().withFillToFit();
- dg.depict(results.get(f).getAc()).writeTo(filename);
- }
- }
-
-
- class ResultComparator implements Comparator
- {
- public int compare(Result o1, Result o2) {
-
- if (o1.getScore() < o2.getScore()) return -1;
- return 1;
- }
- }
-
- private void parseArgs(String[] args) throws ParseException
- {
- Options options = setupOptions(args);
- CommandLineParser parser = new DefaultParser();
- try {
- CommandLine cmd = parser.parse( options, args);
- this.inFile = cmd.getOptionValue("infile");
- this.hoseTSVFile = cmd.getOptionValue("hosecodes");
- this.outPath = cmd.getOptionValue("outpath");
- this.spectrumFile = cmd.getOptionValue("spectrum");
- if (cmd.hasOption("numbers")) this.resultListSize = Integer.parseInt(cmd.getOptionValue("numbers"));
- if (cmd.hasOption("verbose")) this.verbose = true;
- } catch (ParseException e) {
- // TODO Auto-generated catch block
- HelpFormatter formatter = new HelpFormatter();
- formatter.setOptionComparator(null);
- String header = "Ranke structures based on given experimental spectrum and similarity to predicted spectrum.\n\n";
- String footer = "\nPlease report issues at https://github.com/steinbeck/spectra";
- formatter.printHelp( "java -jar casekit.jar casekit.SimilarityRanker", header, options, footer, true );
- throw e;
- }
- }
-
- private Options setupOptions(String[] args)
- {
- Options options = new Options();
-
- Option infile = Option.builder("i")
- .required(true)
- .hasArg()
- .longOpt("infile")
- .desc("filename of with SDF/MOL file of structures to be ranked (required)")
- .build();
- options.addOption(infile);
- Option spectrumfile = Option.builder("p")
- .required(true)
- .hasArg()
- .longOpt("spectrum")
- .desc("filename of CSV file with spectrum. Format of each line: ; (required)")
- .build();
- options.addOption(spectrumfile);
- Option outpath = Option.builder("o")
- .required(true)
- .hasArg()
- .longOpt("outpath")
- .desc("path to store pictures of ranked output structures (required)")
- .build();
- options.addOption(outpath);
- Option hosefile = Option.builder("s")
- .required(true)
- .hasArg()
- .longOpt("hosecodes")
- .desc("filename of TSV file with HOSE codes (required)")
- .build();
- options.addOption(hosefile);
- Option outputnumber = Option.builder("n")
- .hasArg()
- .longOpt("number")
- .desc("number of structures in output file. Default is 10, if this option is ommitted")
- .build();
- options.addOption(outputnumber);
-
- Option verbose = Option.builder("v")
- .required(false)
- .longOpt("verbose")
- .desc("generate messages about progress of operation")
- .build();
- options.addOption(verbose);
-
- return options;
- }
-
-
- public static void main(String[] args) {
- SimilarityRanker sr = new SimilarityRanker();
- try {
- sr.parseArgs(args);
- sr.readSpectrum();
- sr.rank();
- sr.reportResults();
- } catch (Exception e) {
- // TODO Auto-generated catch block
- e.printStackTrace();
- }
-
- }
-
-}
diff --git a/src/casekit/io/FileSystem.java b/src/casekit/io/FileSystem.java
new file mode 100644
index 0000000..63e0e9e
--- /dev/null
+++ b/src/casekit/io/FileSystem.java
@@ -0,0 +1,120 @@
+/*
+ * The MIT License
+ *
+ * Copyright 2019 Michael Wenk [https://github.com/michaelwenk]
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+package casekit.io;
+
+import casekit.nmr.model.DataSet;
+
+import java.io.*;
+import java.nio.file.Files;
+import java.nio.file.Path;
+import java.nio.file.Paths;
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+
+public class FileSystem {
+
+ public static BufferedReader readFile(final String pathToFile) {
+ try {
+ return new BufferedReader(new FileReader(pathToFile));
+ } catch (final IOException e) {
+ e.printStackTrace();
+ }
+
+ return null;
+ }
+
+ public static boolean writeFile(final String pathToFile, final String content) {
+ try {
+ final FileWriter fileWriter = new FileWriter(pathToFile);
+ final BufferedWriter bufferedWriter = new BufferedWriter(fileWriter);
+ bufferedWriter.write(content);
+ bufferedWriter.close();
+
+ return true;
+ } catch (final IOException e) {
+ e.printStackTrace();
+ }
+
+ return false;
+ }
+
+ public static boolean cleanup(final String[] directoriesToCheck, final String pattern) {
+ boolean cleaned = false;
+
+ for (final String dir : directoriesToCheck) {
+ try {
+ cleaned = Files.walk(Paths.get(dir))
+ .map(Path::toFile)
+ .filter(file -> file.getAbsolutePath()
+ .contains(pattern))
+ .allMatch(File::delete);
+
+ } catch (final IOException e) {
+ System.out.println("Not all files could be deleted!");
+ e.printStackTrace();
+ }
+ }
+
+ return cleaned;
+ }
+
+ public static String getFileContent(final String pathToJsonFile) {
+ final BufferedReader bufferedReader = FileSystem.readFile(pathToJsonFile);
+ return bufferedReader
+ == null
+ ? null
+ : bufferedReader.lines()
+ .reduce("", (content, line) -> content
+ + line);
+ }
+
+ public static List getSmilesListFromFile(final String pathToSmilesFile) {
+ final List smilesList = new ArrayList<>();
+ try {
+ final BufferedReader bufferedReader = FileSystem.readFile(pathToSmilesFile);
+ if (bufferedReader
+ != null) {
+ String line;
+ while ((line = bufferedReader.readLine())
+ != null) {
+ smilesList.add(line);
+ }
+ bufferedReader.close();
+ }
+ } catch (final IOException e) {
+ e.printStackTrace();
+ }
+
+ return smilesList;
+ }
+
+ public static List retrieveFromSmilesFile(final String pathToResultsFile) {
+ final List dataSetList = new ArrayList<>();
+ final List smilesList = FileSystem.getSmilesListFromFile(pathToResultsFile);
+
+ DataSet dataSet;
+ Map meta;
+ for (final String smiles : smilesList) {
+ meta = new HashMap<>();
+ meta.put("smiles", smiles);
+ dataSet = new DataSet();
+ dataSet.setMeta(meta);
+
+ dataSetList.add(dataSet);
+ }
+
+ return dataSetList;
+ }
+}
diff --git a/src/casekit/model/NMRSignal.java b/src/casekit/model/NMRSignal.java
deleted file mode 100644
index 5639865..0000000
--- a/src/casekit/model/NMRSignal.java
+++ /dev/null
@@ -1,95 +0,0 @@
-package casekit.model;
-
-/* NMRSignal.java
-*
-* Copyright (C) Dr. Christoph Steinbeck
-*
-* Contact: christoph.steinbeck@uni-jena.de
-*
-* This software is published and distributed under MIT license
-*/
-
-
-/**
-* A class to store the properties of a single N-dimensional NMR signal
-*/
-
-public class NMRSignal {
-
- int dim;
-
- /**
- * Am array of doubles to store the chemical shift of
- */
- public float shift[];
- public String[] nucleus;
-
- /* Signal intensity in arbitrary values */
- public float intensity;
-
- public int phase;
- public static int DIM_ONE = 1, DIM_TWO = 2, DIM_THREE = 3, DIM_FOUR = 4;
- public static int SHIFT_PROTON = 0, SHIFT_HETERO = 1;
- public static int PHASE_NEGATIVE = 2, PHASE_POSITIVE = 1, PHASE_NONE = 0;
- public static String[] PHASENAMES = {"NONE", "POSITIVE", "NEGATIVE"};
-
- public NMRSignal(String[] nucleus) {
- this.dim = nucleus.length;
- this.shift = new float[dim];
- this.nucleus = nucleus;
- for (int f = 0; f < dim; f++)
- shift[f] = 0;
- intensity = 1;
- phase = PHASE_POSITIVE;
- }
-
- public NMRSignal(String[] nucleus, float[] shift, float intensity, int phase) {
- this.dim = nucleus.length;
- this.shift = shift;
- this.nucleus = nucleus;
- this.intensity = intensity;
- this.phase = phase;
- }
-
- public void setShift(float sshift, String nnucleus) {
- for (int f = 0; f < nucleus.length; f++) {
- if (nucleus[f].equals(nnucleus)) {
- shift[f] = sshift;
- break;
- }
- }
- }
-
- public void setShift(float sshift, int dim) {
- shift[dim] = sshift;
- }
-
- public float getShift(String nnucleus) {
- for (int f = 0; f < nucleus.length; f++) {
- if (nucleus[f].equals(nnucleus)) {
- return shift[f];
- }
- }
-
- return Float.MAX_VALUE;
-
- }
-
- public float getShift(int dim) {
- return shift[dim];
- }
-
-
- public String toString() {
- String s = "";
- s += dim + " -dimensional NMRSignal for nuclei ";
- for (int f = 0; f < nucleus.length; f++)
- s += nucleus[f] + "; ";
- s += "\nShiftlist: ";
- for (int f = 0; f < shift.length; f++)
- s += shift[f] + "; ";
- s += "\n\n";
- return s;
- }
-
-}
\ No newline at end of file
diff --git a/src/casekit/model/NMRSpectrum.java b/src/casekit/model/NMRSpectrum.java
deleted file mode 100644
index 27959b2..0000000
--- a/src/casekit/model/NMRSpectrum.java
+++ /dev/null
@@ -1,260 +0,0 @@
-package casekit.model;
-
-/* NMRSpectrum.java
-*
-* Copyright (C) 1997-2007 Christoph Steinbeck
-*
-* Contact: christoph.steinbeck@uni-jena.de
-*
-* This software is published and distributed under MIT License
-*/
-
-/**
-* A Class to model an n-dimensional NMR spectrum,
-*
-*/
-
-import javax.swing.event.ChangeEvent;
-import javax.swing.event.ChangeListener;
-import javax.swing.event.EventListenerList;
-import java.util.ArrayList;
-import java.util.List;
-
-public class NMRSpectrum extends ArrayList{
-
- /**
- * An arbitrary name that can be assigned to this spectrum for identification purposes.
- */
- public String name = "";
- /**
- * An arbitrary name to identify the type of this spectrum, like COSY, NOESY, HSQC, etc. I
- * decided not to provide static Strings with given experiment type since the there are
- * numerous experiments yielding basically identical information having different names
- */
- public String specType = "";
- /**
- * The actual spectrum, i.e. a collection of nmrSignals
- */
- // protected NMRSignal[] nmrSignals;
- /**
- * This holds sorted list of Chemical Shifts of all axes. The first dimension addresses the
- * axes, the second the shift values in this axis, starting from the highest value.
- */
- public List shiftList;
- /**
- * Not yet clear if this is needed.
- */
- public float[] pickPrecision;
- /**
- * Declares how many axes are in involved in this spectrum.
- */
- public int dim = 1;
- /**
- * The nuclei of the different axes.
- */
- public String nucleus[];
- /**
- * The proton frequency of the spectrometer used to record this spectrum.
- */
- public float spectrometerFrequency;
- public String solvent = "";
- public String standard = "";
- /**
- * Some standard nulcei for the 'nucleus' field.
- */
- public static String NUC_PROTON = "1H";
- public static String NUC_CARBON = "13C";
- public static String NUC_NITROGEN = "15N";
- public static String NUC_PHOSPHORUS = "31P";
- // ... to be continued...
- public static String[] SPECTYPE_BB = {NUC_CARBON};
- public static String[] SPECTYPE_DEPT = {NUC_CARBON};
- public static String[] SPECTYPE_HMQC = {NUC_PROTON, NUC_CARBON};
- public static String[] SPECTYPE_HSQC = {NUC_PROTON, NUC_CARBON};
- public static String[] SPECTYPE_NHCORR = {NUC_PROTON, NUC_NITROGEN};
- public static String[] SPECTYPE_HMBC = {NUC_PROTON, NUC_CARBON};
- public static String[] SPECTYPE_HHCOSY = {NUC_PROTON, NUC_PROTON};
- public static String[] SPECTYPE_NOESY = {NUC_PROTON, NUC_PROTON};
- protected transient EventListenerList changeListeners = new EventListenerList();
-
- public NMRSpectrum(String[] nucleus, String name) {
- this.dim = nucleus.length; // redundant, I know :-)
- this.nucleus = nucleus;
- shiftList = new ArrayList(dim);
- for (int f = 0; f < dim; f++) {
- shiftList.add(f, new ArrayList());
- }
- this.name = name;
- }
-
- /**
- * Return the number of individual frequencies in the heteroatom shift list, which should be
- * equal or smaller than the number of respective atoms
- */
- public int getSignalNumber(int axis) {
- return shiftList.get(axis).size();
- }
-
- /**
- * Adds an NMRSignal to the NMRSpectrum.
- */
- public void addSignal(NMRSignal thisSignal) {
- add(thisSignal);
- updateShiftLists();
- }
-
- /**
- * Creates an empty signal with correct dimension
- */
- public void newSignal() {
- System.out.println("nucleus: " + nucleus.length + nucleus[0]);
- add(new NMRSignal(nucleus));
- updateShiftLists();
- }
-
- /**
- * Returns an NMRSignal at position number in the List
- */
- public Object getSignal(int number) {
- return get(number);
- }
-
- /**
- * Returns the position of an NMRSignal the List
- */
- public int getSignalNumber(NMRSignal signal) {
- for (int f = 0; f < size(); f++) {
- if (((NMRSignal) get(f)) == signal) {
- return f;
- }
- }
- return -1;
- }
-
- public void setSpectrometerFrequency(float sf) {
- this.spectrometerFrequency = sf;
- }
-
- public float getSpectrometerFrequency() {
- return spectrometerFrequency;
- }
-
- public void setSolvent(String solvent) {
- this.solvent = solvent;
- }
-
- public String getSolvent() {
- return solvent;
- }
-
- public void setStandard(String standard) {
- this.standard = standard;
- }
-
- public String getStandard() {
- return standard;
- }
-
- /**
- * Returns the signal closest to the shift sought. If no Signal is found within the interval
- * defined by pickprecision, null is returned.
- */
- public Object pickClosestSignal(float shift, String nnucleus,
- float pickprecision) {
- int dim = -1, thisPosition = -1;
- float diff = Float.MAX_VALUE;
- for (int f = 0; f < nucleus.length; f++) {
- if (nucleus[f].equals(nnucleus)) {
- dim = f;
- break;
- }
- }
-
- /*
- * Now we search dimension dim for the chemical shift.
- */
- for (int f = 0; f < size(); f++) {
- if (diff > Math.abs(((NMRSignal) get(f)).shift[dim] - shift)) {
- diff = Math.abs(((NMRSignal) get(f)).shift[dim] - shift);
- diff = (float) Math.ceil(diff * 2) / 2;
- thisPosition = f;
- }
- }
- if (diff < pickprecision) {
- return get(thisPosition);
- }
- return null;
- }
-
- /**
- * Returns a List with signals within the interval defined by pickprecision. If none is found
- * an empty List is returned.
- */
- public List pickSignals(float shift, String nnucleus, float pickprecision) {
- int dim = -1;
- List pickedSignals = new ArrayList();
- for (int f = 0; f < nucleus.length; f++) {
- if (nucleus[f].equals(nnucleus)) {
- dim = f;
- break;
- }
- }
- /*
- * Now we search dimension dim for the chemical shift.
- */
- for (int f = 0; f < size(); f++) {
- if (pickprecision > Math.abs(((NMRSignal) get(f)).shift[dim]
- - shift)) {
- pickedSignals.add(get(f));
- }
- }
- return pickedSignals;
- }
-
- /**
- * Extracts a list of unique shifts from the list of cross signals and sorts them. This is to
- * define the column and row headers for tables.
- */
- protected void updateShiftLists() {
- Float shift;
- for (int i = 0; i < size(); i++) {
- NMRSignal nmrSignal = (NMRSignal) get(i);
- for (int j = 0; j < nmrSignal.shift.length; j++) {
- shift = new Float(nmrSignal.shift[j]);
- if (!shiftList.get(j).contains(shift)) {
- shiftList.get(j).add(shift);
- }
- }
- }
- }
-
- /**
- * Creates a 2D matrix of booleans, that models the set of crosspeaks in the 2D NMR spectrum.
- * The dimensions are taken from hetAtomShiftList and protonShiftList, which again are
- * produced by updateShiftLists based a collection of 2D nmrSignals
- *
- * private void createMatrix(){ boolean found; float het, prot; int hetPos, protPos;
- * hetCorMatrix = new boolean[hetAtomShiftList.length][protonShiftList.length]; for (int f =
- * 0; f < size(); f++){ HetCorNMRSignal hetCorSignal = (HetCorNMRSignal)elementAt(f); prot =
- * hetCorSignal.shift[NMRSignal.SHIFT_PROTON]; het =
- * hetCorSignal.shift[NMRSignal.SHIFT_HETERO]; found = false; hetPos =
- * isInShiftList(hetAtomShiftList, het, hetAtomShiftList.length); if (hetPos >= 0){ protPos =
- * isInShiftList(protonShiftList, prot, protonShiftList.length); if ( protPos >= 0){ found =
- * true; hetCorMatrix[hetPos][protPos] = true; } } } }
- */
- public void report() {
- String s = "";
- System.out.println("Report for nmr spectrum " + name + " of type "
- + specType + ": ");
- for (int i = 0; i < shiftList.size(); i++) {
- System.out.println("ShiftList for dimension " + (i + 1) + ":");
- for (int j = 0; j < shiftList.get(i).size(); j++) {
- s += shiftList.get(i).get(j) + "; ";
- }
- System.out.println(s + "\n");
- s = "";
- }
-
- }
-
-}
diff --git a/src/casekit/nmr/analysis/ConnectivityStatistics.java b/src/casekit/nmr/analysis/ConnectivityStatistics.java
new file mode 100644
index 0000000..d5498fb
--- /dev/null
+++ b/src/casekit/nmr/analysis/ConnectivityStatistics.java
@@ -0,0 +1,423 @@
+package casekit.nmr.analysis;
+
+import casekit.nmr.model.DataSet;
+import casekit.nmr.model.Spectrum;
+import casekit.nmr.utils.Utils;
+import org.openscience.cdk.interfaces.IAtom;
+import org.openscience.cdk.interfaces.IAtomContainer;
+
+import java.util.*;
+import java.util.concurrent.ConcurrentHashMap;
+
+public class ConnectivityStatistics {
+
+ /**
+ * @param dataSet
+ * @param atomType
+ * @param occurrenceStatistics multiplicity -> hybridization -> shift (int) -> elemental composition (mf) -> connected atom symbol -> [#found, #notFound]
+ */
+ public static void buildOccurrenceStatistics(final DataSet dataSet, final String atomType,
+ final Map>>>> occurrenceStatistics) {
+ final Spectrum spectrum = dataSet.getSpectrum()
+ .toSpectrum();
+ final IAtomContainer structure = dataSet.getStructure()
+ .toAtomContainer();
+ final List elements = buildElements(structure);
+ final String elementsString = buildElementsString(elements);
+
+ int shift, atomIndex;
+ IAtom atom;
+ String multiplicity, hybridization;
+ Set found, notFound;
+ for (int signalIndex = 0; signalIndex
+ < spectrum.getSignalCount(); signalIndex++) {
+ shift = spectrum.getShift(signalIndex, 0)
+ .intValue();
+ for (int equivalenceIndex = 0; equivalenceIndex
+ < dataSet.getAssignment()
+ .getAssignment(0, signalIndex).length; equivalenceIndex++) {
+ atomIndex = dataSet.getAssignment()
+ .getAssignment(0, signalIndex, equivalenceIndex);
+ atom = structure.getAtom(atomIndex);
+ if (atom.getSymbol()
+ .equals(atomType)) {
+ multiplicity = Utils.getMultiplicityFromProtonsCount(atom.getImplicitHydrogenCount());
+ if (multiplicity
+ == null) {
+ continue;
+ }
+ multiplicity = multiplicity.toLowerCase();
+ hybridization = atom.getHybridization()
+ .name();
+ occurrenceStatistics.putIfAbsent(multiplicity, new ConcurrentHashMap<>());
+ occurrenceStatistics.get(multiplicity)
+ .putIfAbsent(hybridization, new ConcurrentHashMap<>());
+ occurrenceStatistics.get(multiplicity)
+ .get(hybridization)
+ .putIfAbsent(shift, new ConcurrentHashMap<>());
+ occurrenceStatistics.get(multiplicity)
+ .get(hybridization)
+ .get(shift)
+ .putIfAbsent(elementsString, new ConcurrentHashMap<>());
+ // check for connected hetero atoms
+ found = new HashSet<>();
+ for (final IAtom connectedAtom : structure.getConnectedAtomsList(atom)) {
+ if (connectedAtom.getSymbol()
+ .equals("H")) {
+ continue;
+ }
+ found.add(connectedAtom.getSymbol());
+ }
+ for (final String connectedAtomType : found) {
+ occurrenceStatistics.get(multiplicity)
+ .get(hybridization)
+ .get(shift)
+ .get(elementsString)
+ .putIfAbsent(connectedAtomType, new Integer[]{0, 0});
+ occurrenceStatistics.get(multiplicity)
+ .get(hybridization)
+ .get(shift)
+ .get(elementsString)
+ .get(connectedAtomType)[0] = occurrenceStatistics.get(multiplicity)
+ .get(hybridization)
+ .get(shift)
+ .get(elementsString)
+ .get(connectedAtomType)[0]
+ + 1;
+ }
+ notFound = new HashSet<>(elements);
+ notFound.removeAll(found);
+ for (final String notConnectedAtomType : notFound) {
+ occurrenceStatistics.get(multiplicity)
+ .get(hybridization)
+ .get(shift)
+ .get(elementsString)
+ .putIfAbsent(notConnectedAtomType, new Integer[]{0, 0});
+ occurrenceStatistics.get(multiplicity)
+ .get(hybridization)
+ .get(shift)
+ .get(elementsString)
+ .get(notConnectedAtomType)[1] = occurrenceStatistics.get(multiplicity)
+ .get(hybridization)
+ .get(shift)
+ .get(elementsString)
+ .get(notConnectedAtomType)[1]
+ + 1;
+ }
+ }
+ }
+ }
+ }
+
+ /**
+ * @param structure structure to build and add the statistics from
+ * @param heavyAtomsStatistics elemental composition (mf) -> connected atom pair -> #found
+ */
+ public static void buildHeavyAtomsStatistics(final IAtomContainer structure,
+ final Map> heavyAtomsStatistics) {
+ final List elements = buildElements(structure);
+ final String elementsString = buildElementsString(elements);
+ heavyAtomsStatistics.putIfAbsent(elementsString, new HashMap<>());
+
+ for (final String combination : buildCombinations(elements)) {
+ heavyAtomsStatistics.get(elementsString)
+ .putIfAbsent(combination, 0);
+ }
+
+ IAtom atom;
+ String atomPairKey;
+ final Map found = new HashMap<>();
+ for (int i = 0; i
+ < structure.getAtomCount(); i++) {
+ atom = structure.getAtom(i);
+ // check for connected hetero atoms
+ for (final IAtom connectedAtom : structure.getConnectedAtomsList(atom)) {
+ if (connectedAtom.getSymbol()
+ .equals("H")) {
+ continue;
+ }
+ atomPairKey = buildAtomPairString(atom.getSymbol(), connectedAtom.getSymbol());
+ found.putIfAbsent(atomPairKey, 0);
+ found.put(atomPairKey, found.get(atomPairKey)
+ + 1);
+ }
+ }
+ for (final String connectedAtomPair : found.keySet()) {
+ heavyAtomsStatistics.get(elementsString)
+ .put(connectedAtomPair, heavyAtomsStatistics.get(elementsString)
+ .get(connectedAtomPair)
+ + found.get(connectedAtomPair)
+ / 2); // divided by two since we count for both bond partners in the previous loop
+ }
+ }
+
+ public static List buildElements(final IAtomContainer structure) {
+ final String mf = Utils.molecularFormularToString(Utils.getMolecularFormulaFromAtomContainer(structure));
+ return buildElements(mf);
+ }
+
+ public static List buildElements(final String mf) {
+ final List elements = new ArrayList<>(Utils.getMolecularFormulaElementCounts(mf)
+ .keySet());
+ elements.remove("H");
+ Collections.sort(elements);
+
+ return elements;
+ }
+
+ public static String buildElementsString(final List elements) {
+ return String.join(",", elements);
+ }
+
+ public static String buildAtomPairString(final String atomType1, final String atomType2) {
+ final List atomPairList = new ArrayList<>();
+ atomPairList.add(atomType1);
+ atomPairList.add(atomType2);
+ Collections.sort(atomPairList);
+
+ return String.join("_", atomPairList);
+ }
+
+ public static Set buildCombinations(final List elements) {
+ final Set combinations = new HashSet<>();
+ for (final String element1 : elements) {
+ for (final String element2 : elements) {
+ combinations.add(buildAtomPairString(element1, element2));
+ }
+ }
+
+ return combinations;
+ }
+
+ // /**
+ // * @param dataSetList
+ // * @param nucleus
+ // * @param connectivityStatistics multiplicity -> hybridization -> shift (int) -> connected atom symbol -> connected atom hybridization -> connected atom protons count -> occurrence
+ // */
+ // @Deprecated
+ // public static void buildConnectivityStatistics(final List dataSetList, final String nucleus,
+ // final Map>>>>> connectivityStatistics) {
+ // final String atomType = Utils.getAtomTypeFromNucleus(nucleus);
+ // for (final DataSet dataSet : dataSetList) {
+ // if (!dataSet.getSpectrum()
+ // .getNuclei()[0].equals(nucleus)) {
+ // continue;
+ // }
+ // buildConnectivityStatistics(dataSet, atomType, connectivityStatistics);
+ // }
+ // }
+ //
+ // /**
+ // * @param dataSet
+ // * @param atomType
+ // * @param connectivityStatistics multiplicity -> hybridization -> shift (int) -> connected atom symbol -> connected atom hybridization -> connected atom protons count -> occurrence
+ // */
+ // @Deprecated
+ // public static void buildConnectivityStatistics(final DataSet dataSet, final String atomType,
+ // final Map>>>>> connectivityStatistics) {
+ // final IAtomContainer structure = dataSet.getStructure()
+ // .toAtomContainer();
+ // final Spectrum spectrum = dataSet.getSpectrum()
+ // .toSpectrum();
+ // int shift, atomIndex;
+ // IAtom atom;
+ // String multiplicity, hybridization, connectedAtomType, connectedAtomHybridization;
+ // for (int signalIndex = 0; signalIndex
+ // < spectrum.getSignalCount(); signalIndex++) {
+ // shift = spectrum.getShift(signalIndex, 0)
+ // .intValue();
+ // for (int equivalenceIndex = 0; equivalenceIndex
+ // < dataSet.getAssignment()
+ // .getAssignment(0, signalIndex).length; equivalenceIndex++) {
+ // atomIndex = dataSet.getAssignment()
+ // .getAssignment(0, signalIndex, equivalenceIndex);
+ // atom = structure.getAtom(atomIndex);
+ // if (atom.getSymbol()
+ // .equals(atomType)) {
+ // multiplicity = Utils.getMultiplicityFromProtonsCount(atom.getImplicitHydrogenCount());
+ // if (multiplicity
+ // == null) {
+ // continue;
+ // }
+ // multiplicity = multiplicity.toLowerCase();
+ // hybridization = atom.getHybridization()
+ // .name();
+ // connectivityStatistics.putIfAbsent(multiplicity, new ConcurrentHashMap<>());
+ // connectivityStatistics.get(multiplicity)
+ // .putIfAbsent(hybridization, new ConcurrentHashMap<>());
+ // // check for connected hetero atoms
+ // for (final IAtom connectedAtom : structure.getConnectedAtomsList(atom)) {
+ // if (connectedAtom.getSymbol()
+ // .equals("H")) {
+ // continue;
+ // }
+ // connectedAtomType = connectedAtom.getSymbol();
+ // if (connectedAtom.getHybridization()
+ // == null) {
+ // continue;
+ // }
+ // connectedAtomHybridization = connectedAtom.getHybridization()
+ // .name();
+ // connectivityStatistics.get(multiplicity)
+ // .get(hybridization)
+ // .putIfAbsent(shift, new ConcurrentHashMap<>());
+ // connectivityStatistics.get(multiplicity)
+ // .get(hybridization)
+ // .get(shift)
+ // .putIfAbsent(connectedAtomType, new ConcurrentHashMap<>());
+ // connectivityStatistics.get(multiplicity)
+ // .get(hybridization)
+ // .get(shift)
+ // .get(connectedAtomType)
+ // .putIfAbsent(connectedAtomHybridization, new ConcurrentHashMap<>());
+ // connectivityStatistics.get(multiplicity)
+ // .get(hybridization)
+ // .get(shift)
+ // .get(connectedAtomType)
+ // .get(connectedAtomHybridization)
+ // .putIfAbsent(connectedAtom.getImplicitHydrogenCount(), 0);
+ // connectivityStatistics.get(multiplicity)
+ // .get(hybridization)
+ // .get(shift)
+ // .get(connectedAtomType)
+ // .get(connectedAtomHybridization)
+ // .put(connectedAtom.getImplicitHydrogenCount(), connectivityStatistics.get(
+ // multiplicity)
+ // .get(hybridization)
+ // .get(shift)
+ // .get(connectedAtomType)
+ // .get(connectedAtomHybridization)
+ // .get(connectedAtom.getImplicitHydrogenCount())
+ // + 1);
+ // }
+ // }
+ // }
+ // }
+ // }
+ //
+ // /**
+ // * @param connectivityStatistics multiplicity -> hybridization -> shift (int) -> connected atom symbol -> connected atom hybridization -> connected atom protons count -> occurrence
+ // * @param multiplicity
+ // * @param hybridization
+ // * @param shift
+ // * @param molecularFormulaElements
+ // *
+ // * @return
+ // */
+ // @Deprecated
+ // public static Map>> extractConnectivities(
+ // final Map>>>>> connectivityStatistics,
+ // final String multiplicity, final String hybridization, final int shift,
+ // final Set molecularFormulaElements) {
+ // final Map>> extractedConnectivities = new HashMap<>();
+ // if (connectivityStatistics.containsKey(multiplicity)
+ // && connectivityStatistics.get(multiplicity)
+ // .containsKey(hybridization)
+ // && connectivityStatistics.get(multiplicity)
+ // .get(hybridization)
+ // .containsKey(shift)) {
+ // for (final Map.Entry>> entry : connectivityStatistics.get(
+ // multiplicity)
+ // .get(hybridization)
+ // .get(shift)
+ // .entrySet()) {
+ // if (molecularFormulaElements.contains(entry.getKey())) {
+ // extractedConnectivities.put(entry.getKey(), entry.getValue());
+ // }
+ // }
+ // }
+ //
+ // return extractedConnectivities;
+ // }
+ //
+ // @Deprecated
+ // public static Map>> filterExtractedConnectivitiesByHybridizations(
+ // final Map>> extractedConnectivities,
+ // final Set knownCarbonHybridizations) {
+ // // remove hybridization of carbons which we do not expect
+ // for (final String atomType : extractedConnectivities.keySet()) {
+ // if (atomType.equals("C")) {
+ // for (final int hybridization : new HashSet<>(extractedConnectivities.get(atomType)
+ // .keySet())) {
+ // if (!knownCarbonHybridizations.contains(hybridization)) {
+ // extractedConnectivities.get(atomType)
+ // .remove(hybridization);
+ // }
+ // }
+ // }
+ // }
+ //
+ // return extractedConnectivities;
+ // }
+ //
+ // @Deprecated
+ // public static Map>> filterExtractedConnectivitiesByCount(
+ // final Map>> extractedConnectivities,
+ // final double thresholdElementCount, final boolean onAtomTypeLevel) {
+ // final Map totalCounts = getTotalCounts(extractedConnectivities);
+ // final int totalCountsSum = getSum(new HashSet<>(totalCounts.values()));
+ // final Map>> filteredExtractedConnectivities = new HashMap<>();
+ // extractedConnectivities.keySet()
+ // .forEach(neighborAtomType -> {
+ // int sum = 0;
+ // for (final Map.Entry> entryPerHybridization : extractedConnectivities.get(
+ // neighborAtomType)
+ // .entrySet()) {
+ // for (final Map.Entry entryProtonsCount : extractedConnectivities.get(
+ // neighborAtomType)
+ // .get(entryPerHybridization.getKey())
+ // .entrySet()) {
+ // if (onAtomTypeLevel) {
+ // sum += entryProtonsCount.getValue();
+ // } else if (entryProtonsCount.getValue()
+ // / (double) totalCountsSum
+ // >= thresholdElementCount) {
+ // filteredExtractedConnectivities.putIfAbsent(neighborAtomType,
+ // new HashMap<>());
+ // filteredExtractedConnectivities.get(neighborAtomType)
+ // .putIfAbsent(
+ // entryPerHybridization.getKey(),
+ // new HashSet<>());
+ // filteredExtractedConnectivities.get(neighborAtomType)
+ // .get(entryPerHybridization.getKey())
+ // .add(entryProtonsCount.getKey());
+ // }
+ // }
+ // }
+ // if (onAtomTypeLevel
+ // && sum
+ // / (double) totalCountsSum
+ // >= thresholdElementCount) {
+ // filteredExtractedConnectivities.putIfAbsent(neighborAtomType, new HashMap<>());
+ // }
+ // });
+ //
+ // return filteredExtractedConnectivities;
+ // }
+ //
+ // @Deprecated
+ // private static Map getTotalCounts(
+ // final Map>> extractedConnectivities) {
+ // final Map totalCounts = new HashMap<>();
+ // for (final String key1 : extractedConnectivities.keySet()) {
+ // totalCounts.putIfAbsent(key1, 0);
+ // for (final int key2 : extractedConnectivities.get(key1)
+ // .keySet()) {
+ // for (final Map.Entry countsEntry : extractedConnectivities.get(key1)
+ // .get(key2)
+ // .entrySet()) {
+ // totalCounts.put(key1, totalCounts.get(key1)
+ // + countsEntry.getValue());
+ // }
+ // }
+ // }
+ //
+ // return totalCounts;
+ // }
+ //
+ // @Deprecated
+ // private static int getSum(final Set values) {
+ // return values.stream()
+ // .reduce(0, (total, current) -> total += current);
+ // }
+}
diff --git a/src/casekit/nmr/analysis/HOSECodeShiftStatistics.java b/src/casekit/nmr/analysis/HOSECodeShiftStatistics.java
new file mode 100644
index 0000000..e0fecad
--- /dev/null
+++ b/src/casekit/nmr/analysis/HOSECodeShiftStatistics.java
@@ -0,0 +1,318 @@
+package casekit.nmr.analysis;
+
+import casekit.nmr.dbservice.COCONUT;
+import casekit.nmr.dbservice.NMRShiftDB;
+import casekit.nmr.fragments.model.ConnectionTree;
+import casekit.nmr.hose.HOSECodeBuilder;
+import casekit.nmr.model.DataSet;
+import casekit.nmr.model.Signal;
+import casekit.nmr.model.Spectrum;
+import casekit.nmr.utils.Statistics;
+import casekit.nmr.utils.Utils;
+import com.google.gson.Gson;
+import com.google.gson.GsonBuilder;
+import com.google.gson.JsonObject;
+import com.google.gson.JsonParser;
+import com.google.gson.reflect.TypeToken;
+import org.bson.Document;
+import org.openscience.cdk.exception.CDKException;
+import org.openscience.cdk.interfaces.IAtomContainer;
+import org.openscience.nmrshiftdb.util.ExtendedHOSECodeGenerator;
+
+import java.io.*;
+import java.util.*;
+
+public class HOSECodeShiftStatistics {
+
+ private final static Gson GSON = new GsonBuilder().setLenient()
+ .create();
+ private final static ExtendedHOSECodeGenerator extendedHOSECodeGenerator = new ExtendedHOSECodeGenerator();
+
+ public static Map>> collectHOSECodeShifts(final List dataSetList,
+ final Integer maxSphere,
+ final boolean use3D,
+ final boolean withExplicitH) {
+ return collectHOSECodeShifts(dataSetList, maxSphere, use3D, withExplicitH, new HashMap<>());
+ }
+
+ /**
+ * This method expects datasets containing structures without explicit hydrogens.
+ *
+ * @param dataSetList
+ * @param maxSphere
+ * @param hoseCodeShifts
+ *
+ * @return
+ */
+ public static Map>> collectHOSECodeShifts(final List dataSetList,
+ final Integer maxSphere,
+ final boolean use3D,
+ final boolean withExplicitH,
+ final Map>> hoseCodeShifts) {
+ for (final DataSet dataSet : dataSetList) {
+ insert(dataSet, maxSphere, use3D, withExplicitH, hoseCodeShifts);
+ }
+
+ return hoseCodeShifts;
+ }
+
+ public static boolean insert(final DataSet dataSet, final Integer maxSphere, final boolean use3D,
+ final boolean withExplicitH,
+ final Map>> hoseCodeShifts) {
+ final IAtomContainer structure;
+ Signal signal;
+ String hoseCode;
+ final String atomTypeSpectrum;
+ String solvent;
+ final Map atomIndexMap; // from explicit H to heavy atom
+ ConnectionTree connectionTree;
+ int maxSphereTemp;
+ List signalIndices;
+ structure = dataSet.getStructure()
+ .toAtomContainer();
+ final Spectrum spectrum = dataSet.getSpectrum()
+ .toSpectrum();
+ if (Utils.containsExplicitHydrogens(structure)) {
+ System.out.println("!!!Dataset skipped must not contain (previously set) explicit hydrogens!!!");
+ return false;
+ }
+ // create atom index map to know which indices the explicit hydrogens will have
+ atomIndexMap = new HashMap<>();
+ if (use3D
+ || withExplicitH) {
+ try {
+ int nextAtomIndexExplicitH = structure.getAtomCount();
+ for (int i = 0; i
+ < structure.getAtomCount(); i++) {
+ if (structure.getAtom(i)
+ .getImplicitHydrogenCount()
+ != null) {
+ for (int j = 0; j
+ < structure.getAtom(i)
+ .getImplicitHydrogenCount(); j++) {
+ atomIndexMap.put(nextAtomIndexExplicitH, i);
+ nextAtomIndexExplicitH++;
+ }
+ }
+ }
+
+ if (use3D) {
+ try {
+ Utils.placeExplicitHydrogens(structure);
+ } catch (final CDKException | IOException | ClassNotFoundException e) {
+ e.printStackTrace();
+ }
+ } else {
+ Utils.convertImplicitToExplicitHydrogens(structure);
+ }
+ Utils.setAromaticityAndKekulize(structure);
+ } catch (final CDKException e) {
+ e.printStackTrace();
+ return false;
+ }
+ }
+ solvent = dataSet.getSpectrum()
+ .getMeta()
+ == null
+ ? null
+ : dataSet.getSpectrum()
+ .getMeta()
+ .get("solvent");
+ if (solvent
+ == null
+ || solvent.equals("")) {
+ solvent = "Unknown";
+ }
+ atomTypeSpectrum = Utils.getAtomTypeFromNucleus(dataSet.getSpectrum()
+ .getNuclei()[0]);
+ for (int i = 0; i
+ < structure.getAtomCount(); i++) {
+ signalIndices = null;
+ if (structure.getAtom(i)
+ .getSymbol()
+ .equals(atomTypeSpectrum)) {
+ if (atomTypeSpectrum.equals("H")) {
+ // could be multiple signals
+ signalIndices = dataSet.getAssignment()
+ .getIndices(0, atomIndexMap.get(i));
+ } else {
+ // should be one only
+ signalIndices = dataSet.getAssignment()
+ .getIndices(0, i);
+ }
+ }
+ if (signalIndices
+ != null) {
+ for (final Integer signalIndex : signalIndices) {
+ signal = spectrum.getSignal(signalIndex);
+ try {
+ if (maxSphere
+ == null) {
+ connectionTree = HOSECodeBuilder.buildConnectionTree(structure, i, null);
+ maxSphereTemp = connectionTree.getMaxSphere(true);
+ } else {
+ maxSphereTemp = maxSphere;
+ }
+ for (int sphere = 1; sphere
+ <= maxSphereTemp; sphere++) {
+ if (use3D) {
+ try {
+ hoseCode = extendedHOSECodeGenerator.getHOSECode(structure, structure.getAtom(i),
+ sphere);
+ } catch (final Exception e) {
+ // e.printStackTrace();
+ continue;
+ }
+ } else {
+ hoseCode = HOSECodeBuilder.buildHOSECode(structure, i, sphere, false);
+ }
+ hoseCodeShifts.putIfAbsent(hoseCode, new HashMap<>());
+ hoseCodeShifts.get(hoseCode)
+ .putIfAbsent(solvent, new ArrayList<>());
+ hoseCodeShifts.get(hoseCode)
+ .get(solvent)
+ .add(signal.getShift(0));
+ }
+ } catch (final CDKException e) {
+ e.printStackTrace();
+ }
+ }
+ }
+ }
+
+ return true;
+ }
+
+ public static Map> buildHOSECodeShiftStatistics(
+ final Map>> hoseCodeShifts) {
+
+ final Map> hoseCodeShiftStatistics = new HashMap<>();
+ List values;
+ for (final Map.Entry>> hoseCodes : hoseCodeShifts.entrySet()) {
+ hoseCodeShiftStatistics.put(hoseCodes.getKey(), new HashMap<>());
+ for (final Map.Entry> solvents : hoseCodes.getValue()
+ .entrySet()) {
+ values = new ArrayList<>(solvents.getValue());
+ values = Statistics.removeOutliers(values, 1.5);
+ hoseCodeShiftStatistics.get(hoseCodes.getKey())
+ .put(solvents.getKey(),
+ new Double[]{(double) values.size(), Collections.min(values),
+ Statistics.getMean(values), Statistics.getMedian(values),
+ Collections.max(values)});
+ }
+ }
+
+ return hoseCodeShiftStatistics;
+ }
+
+ public static Map> buildHOSECodeShiftStatistics(final String[] pathsToNMRShiftDBs,
+ final String[] pathsToCOCONUTs,
+ final String[] nuclei,
+ final Integer maxSphere,
+ final boolean use3D,
+ final boolean withExplicitH) {
+ try {
+ final Map>> hoseCodeShifts = new HashMap<>();
+ for (final String pathsToNMRShiftDB : pathsToNMRShiftDBs) {
+ HOSECodeShiftStatistics.collectHOSECodeShifts(
+ NMRShiftDB.getDataSetsFromNMRShiftDB(pathsToNMRShiftDB, nuclei), maxSphere, use3D,
+ withExplicitH, hoseCodeShifts);
+ }
+ for (final String pathsToCOCONUT : pathsToCOCONUTs) {
+ HOSECodeShiftStatistics.collectHOSECodeShifts(
+ COCONUT.getDataSetsWithShiftPredictionFromCOCONUT(pathsToCOCONUT, nuclei), maxSphere, use3D,
+ withExplicitH, hoseCodeShifts);
+ }
+ return HOSECodeShiftStatistics.buildHOSECodeShiftStatistics(hoseCodeShifts);
+ } catch (final FileNotFoundException | CDKException e) {
+ e.printStackTrace();
+ }
+
+ return new HashMap<>();
+ }
+
+ public static Map> buildHOSECodeShiftStatistics(final List dataSetList,
+ final Integer maxSphere,
+ final boolean use3D,
+ final boolean withExplicitH) {
+ return HOSECodeShiftStatistics.buildHOSECodeShiftStatistics(
+ collectHOSECodeShifts(dataSetList, maxSphere, use3D, withExplicitH));
+ }
+
+ public static boolean writeHOSECodeShiftStatistics(final Map> hoseCodeShifts,
+ final String pathToJsonFile) {
+ try {
+ final BufferedWriter bw = new BufferedWriter(new FileWriter(pathToJsonFile));
+ bw.append("{");
+ bw.newLine();
+ bw.flush();
+
+ Document subDocument;
+ String json;
+ long counter = 0;
+ for (final Map.Entry> entry : hoseCodeShifts.entrySet()) {
+ subDocument = new Document();
+ subDocument.append("HOSECode", entry.getKey());
+ subDocument.append("values", GSON.toJson(entry.getValue()));
+ json = new Document(String.valueOf(counter), subDocument).toJson();
+ bw.append(json, 1, json.length()
+ - 1);
+ if (counter
+ < hoseCodeShifts.size()
+ - 1) {
+ bw.append(",");
+ }
+ bw.newLine();
+ bw.flush();
+
+ counter++;
+ }
+
+ bw.append("}");
+ bw.flush();
+ bw.close();
+
+ return true;
+ } catch (final IOException e) {
+ e.printStackTrace();
+ }
+
+ return false;
+ }
+
+ public static Map> readHOSECodeShiftStatistics(
+ final String pathToJsonFile) throws FileNotFoundException {
+ final BufferedReader br = new BufferedReader(new FileReader(pathToJsonFile));
+ final Map> hoseCodeShiftStatistics = new HashMap<>();
+ // add all task to do
+ br.lines()
+ .forEach(line -> {
+ if ((line.trim()
+ .length()
+ > 1)
+ || (!line.trim()
+ .startsWith("{")
+ && !line.trim()
+ .endsWith("}"))) {
+ final StringBuilder hoseCodeShiftsStatisticInJSON = new StringBuilder();
+ if (line.endsWith(",")) {
+ hoseCodeShiftsStatisticInJSON.append(line, 0, line.length()
+ - 1);
+ } else {
+ hoseCodeShiftsStatisticInJSON.append(line);
+ }
+ final JsonObject jsonObject = JsonParser.parseString(hoseCodeShiftsStatisticInJSON.substring(
+ hoseCodeShiftsStatisticInJSON.toString()
+ .indexOf("{")))
+ .getAsJsonObject();
+ hoseCodeShiftStatistics.put(jsonObject.get("HOSECode")
+ .getAsString(), GSON.fromJson(jsonObject.get("values")
+ .getAsString(),
+ new TypeToken>() {
+ }.getType()));
+ }
+ });
+
+ return hoseCodeShiftStatistics;
+ }
+}
diff --git a/src/casekit/nmr/analysis/MultiplicitySectionsBuilder.java b/src/casekit/nmr/analysis/MultiplicitySectionsBuilder.java
new file mode 100644
index 0000000..04dc01d
--- /dev/null
+++ b/src/casekit/nmr/analysis/MultiplicitySectionsBuilder.java
@@ -0,0 +1,170 @@
+/*
+ * The MIT License
+ *
+ * Copyright (c) 2019 Michael Wenk [https://github.com/michaelwenk]
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+package casekit.nmr.analysis;
+
+import casekit.nmr.model.Signal;
+import casekit.nmr.model.Spectrum;
+
+import java.util.*;
+
+/**
+ * @author Michael Wenk [https://github.com/michaelwenk]
+ */
+public class MultiplicitySectionsBuilder {
+
+ private final HashSet multiplicities;
+ private int minLimit, maxLimit, stepSize, steps;
+
+ public MultiplicitySectionsBuilder() {
+ this.multiplicities = new HashSet<>();
+ this.init();
+ }
+
+ private void init() {
+ this.multiplicities.clear();
+ this.multiplicities.add("s");
+ this.multiplicities.add("d");
+ this.multiplicities.add("t");
+ this.multiplicities.add("q");
+ this.multiplicities.add("unknown");
+ this.minLimit = -50;
+ this.maxLimit = 300;
+ this.stepSize = 5;
+ this.updateSteps(); // ppm range from -20 to 260 in 5 ppm steps
+ }
+
+ /**
+ * Resets to following default values:
+ * multiplicties: S, D, T. Q
+ * min. ppm limit: -50
+ * max. ppm limit: 300
+ * step size: 5
+ */
+ public void reset() {
+ this.init();
+ }
+
+ public Map> buildMultiplicitySections(final Spectrum spectrum, final int dim) {
+ final Map> multiplicitySections = new HashMap<>();
+ // init
+ for (final String multiplicity : this.multiplicities) {
+ multiplicitySections.put(multiplicity, new ArrayList<>());
+ }
+ // set the mult. sections
+ Signal signal;
+ Integer shiftSection;
+ String multiplicity;
+ for (int i = 0; i
+ < spectrum.getSignalCount(); i++) {
+ signal = spectrum.getSignal(i);
+ shiftSection = this.calculateShiftSection(signal, dim);
+ if (shiftSection
+ == null) {
+ System.err.println("MultiplicitySectionsBuilder: signal or its chemical shift is missing: "
+ + signal);
+ continue;
+ }
+ multiplicity = this.checkMultiplicity(signal);
+ if (multiplicity
+ == null) {
+ System.err.println("MultiplicitySectionsBuilder: signal multiplicity is not in list: "
+ + signal);
+ continue;
+ }
+ multiplicitySections.get(multiplicity)
+ .add(shiftSection);
+ }
+
+ return multiplicitySections;
+ }
+
+ public Integer calculateShiftSection(final Signal signal, final int dim) {
+ if (signal
+ == null
+ || signal.getShift(dim)
+ == null) {
+ return null;
+ }
+ return (int) ((signal.getShift(dim)
+ - this.minLimit)
+ / this.stepSize);
+ }
+
+ public String checkMultiplicity(final Signal signal) {
+ final String multiplicity = signal.getMultiplicity()
+ != null
+ ? signal.getMultiplicity()
+ : "unknown";
+ if (!this.multiplicities.contains(multiplicity)) {
+ return null;
+ }
+
+ return multiplicity;
+ }
+
+ public Set getMultiplicities() {
+ return this.multiplicities;
+ }
+
+ public boolean addMultiplicity(final String mult) {
+ return this.multiplicities.add(mult);
+ }
+
+ public boolean removeMultiplicity(final String mult) {
+ return this.multiplicities.remove(mult);
+ }
+
+ public boolean containsMultiplicity(final String mult) {
+ return this.multiplicities.contains(mult);
+ }
+
+ public int getMinLimit() {
+ return this.minLimit;
+ }
+
+ public void setMinLimit(final int minLimit) {
+ this.minLimit = minLimit;
+ this.updateSteps();
+ }
+
+ public int getMaxLimit() {
+ return this.maxLimit;
+ }
+
+ public void setMaxLimit(final int maxLimit) {
+ this.maxLimit = maxLimit;
+ this.updateSteps();
+ }
+
+ public int getStepSize() {
+ return this.stepSize;
+ }
+
+ public void setStepSize(final int stepSize) {
+ this.stepSize = stepSize;
+ this.updateSteps();
+ }
+
+ public int getSteps() {
+ return this.steps;
+ }
+
+ private void updateSteps() {
+ this.steps = this.calculateSteps(this.minLimit, this.maxLimit, this.stepSize);
+ }
+
+ public int calculateSteps(final int minLimit, final int maxLimit, final int stepSize) {
+ return (maxLimit
+ - minLimit)
+ / stepSize;
+ }
+}
diff --git a/src/casekit/nmr/dbservice/COCONUT.java b/src/casekit/nmr/dbservice/COCONUT.java
new file mode 100644
index 0000000..577678e
--- /dev/null
+++ b/src/casekit/nmr/dbservice/COCONUT.java
@@ -0,0 +1,102 @@
+package casekit.nmr.dbservice;
+
+import casekit.nmr.model.*;
+import casekit.nmr.utils.Utils;
+import org.openscience.cdk.exception.CDKException;
+import org.openscience.cdk.interfaces.IAtomContainer;
+import org.openscience.cdk.io.iterator.IteratingSDFReader;
+import org.openscience.cdk.silent.SilentChemObjectBuilder;
+
+import java.io.FileNotFoundException;
+import java.io.FileReader;
+import java.util.ArrayList;
+import java.util.List;
+
+public class COCONUT {
+
+ public static List getDataSetsWithShiftPredictionFromCOCONUT(final String pathToCOCONUT,
+ final String[] nuclei) throws CDKException, FileNotFoundException {
+ final List dataSetList = new ArrayList<>();
+ final IteratingSDFReader iterator = new IteratingSDFReader(new FileReader(pathToCOCONUT),
+ SilentChemObjectBuilder.getInstance());
+ IAtomContainer structure;
+ DataSet dataSet;
+ Spectrum spectrum;
+ Assignment assignment;
+ String[] split, split2;
+ String spectrumPropertyString, multiplicity;
+ double calcShift;
+ List closestSignalList;
+ int atomIndex;
+
+ while (iterator.hasNext()) {
+ structure = iterator.next();
+ dataSet = Utils.atomContainerToDataSet(structure);
+
+ for (final String nucleus : nuclei) {
+ final String atomType = casekit.nmr.utils.Utils.getAtomTypeFromNucleus(nucleus);
+ final List atomIndices = Utils.getAtomTypeIndicesByElement(structure, atomType);
+ spectrumPropertyString = structure.getProperty("Predicted "
+ + nucleus
+ + " shifts", String.class);
+ if (spectrumPropertyString
+ == null) {
+ spectrumPropertyString = structure.getProperty("Predicted_"
+ + nucleus
+ + "_shifts", String.class);
+ }
+ if (spectrumPropertyString
+ == null) {
+ continue;
+ }
+
+ spectrumPropertyString = spectrumPropertyString.replaceAll("[\\n\\r]", ";");
+ split = spectrumPropertyString.split(";");
+ spectrum = new Spectrum();
+ spectrum.setNuclei(new String[]{nucleus});
+ spectrum.setSignals(new ArrayList<>());
+ assignment = new Assignment();
+ assignment.setNuclei(spectrum.getNuclei());
+ assignment.initAssignments(spectrum.getSignalCount());
+ for (int i = 0; i
+ < split.length; i++) {
+ split2 = split[i].split("\\s+");
+ atomIndex = atomIndices.get(i);
+ calcShift = Double.parseDouble(split2[1]);
+ multiplicity = Utils.getMultiplicityFromProtonsCount(structure.getAtom(atomIndex)
+ .getImplicitHydrogenCount())
+ .toLowerCase();
+ // add assignment (at first here because of search for already existing equivalent signals)
+ // just to be sure that we take the right signal if equivalences are present
+ closestSignalList = spectrum.pickByClosestShift(calcShift, 0, 0.0);
+ closestSignalList.retainAll(spectrum.pickByMultiplicity(multiplicity));
+ if (closestSignalList.isEmpty()) {
+ assignment.addAssignment(0, new int[]{atomIndex});
+ } else {
+ assignment.addAssignmentEquivalence(0, closestSignalList.get(0), atomIndex);
+ }
+ // add signal
+ spectrum.addSignal(
+ new Signal(new String[]{nucleus}, new Double[]{calcShift}, multiplicity, "signal", null, 1,
+ 0, null, null));
+ }
+
+ // if no spectrum could be built or the number of signals in spectrum is different than the atom number in molecule
+ if (Utils.getDifferenceSpectrumSizeAndMolecularFormulaCount(spectrum,
+ Utils.getMolecularFormulaFromString(
+ dataSet.getMeta()
+ .get("mf")), 0)
+ != 0) {
+ continue;
+ }
+ dataSet.setSpectrum(new SpectrumCompact(spectrum));
+ dataSet.setAssignment(assignment);
+
+ dataSetList.add(dataSet.buildClone());
+ }
+ }
+
+ return dataSetList;
+ }
+
+}
diff --git a/src/casekit/nmr/dbservice/NMRShiftDB.java b/src/casekit/nmr/dbservice/NMRShiftDB.java
new file mode 100644
index 0000000..5c2ae4a
--- /dev/null
+++ b/src/casekit/nmr/dbservice/NMRShiftDB.java
@@ -0,0 +1,398 @@
+/*
+ * The MIT License
+ *
+ * Copyright 2019 Michael Wenk [https://github.com/michaelwenk]
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+package casekit.nmr.dbservice;
+
+import casekit.nmr.model.*;
+import casekit.nmr.utils.Utils;
+import org.openscience.cdk.exception.CDKException;
+import org.openscience.cdk.interfaces.IAtomContainer;
+import org.openscience.cdk.io.iterator.IteratingSDFReader;
+import org.openscience.cdk.silent.SilentChemObjectBuilder;
+import org.openscience.cdk.tools.manipulator.AtomContainerManipulator;
+
+import java.io.FileNotFoundException;
+import java.io.FileReader;
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.List;
+
+public class NMRShiftDB {
+
+ public static String getSolvent(final String solventPropertyString, final String spectrumIndexInRecord) {
+ final String[] solventPropertyStringSplit = solventPropertyString.split(":");
+ String solvent;
+ for (int i = 0; i
+ < solventPropertyStringSplit.length; i++) {
+ if (solventPropertyStringSplit[i].endsWith(spectrumIndexInRecord)) {
+ solvent = solventPropertyStringSplit[i
+ + 1];
+ if (solvent.substring(solvent.length()
+ - 1)
+ .matches("\\d")) {
+ solvent = solvent.substring(0, solvent.length()
+ - 1);
+ }
+ if (solvent.substring(solvent.length()
+ - 1)
+ .matches("\\d")) {
+ solvent = solvent.substring(0, solvent.length()
+ - 1);
+ }
+ // solvent = solvent.substring(0, solvent.length() - 1);
+ solvent = solvent.trim();
+
+ return solvent;
+ }
+ }
+
+ return null;
+ }
+
+ public static List getSpectraProperties1D(final IAtomContainer ac, final String nucleus) {
+ final List spectraProperties1D = new ArrayList<>();
+ for (final Object obj : ac.getProperties()
+ .keySet()) {
+ if (obj instanceof String
+ && ((String) obj).startsWith("Spectrum "
+ + nucleus)) {
+ spectraProperties1D.add((String) obj);
+ }
+ }
+
+ return spectraProperties1D;
+ }
+
+ /**
+ * Returns a {@link DataSet} class object
+ * for each valid molecule record in the given NMRShiftDB file. Valid means
+ * here that each molecule record has to contain the given spectrum
+ * property string as well as the number of signals in that spectrum has to
+ * be the same as atoms of that atom type in molecule.
+ *
+ * @param pathToNMRShiftDB path to NMRShiftDB file
+ * @param nuclei nuclei to get the spectra for
+ *
+ * @return
+ *
+ * @throws FileNotFoundException
+ * @throws CDKException
+ * @see DataSet
+ */
+ public static List getDataSetsFromNMRShiftDB(final String pathToNMRShiftDB,
+ final String[] nuclei) throws FileNotFoundException, CDKException {
+ final List dataSets = new ArrayList<>();
+ final IteratingSDFReader iterator = new IteratingSDFReader(new FileReader(pathToNMRShiftDB),
+ SilentChemObjectBuilder.getInstance());
+ IAtomContainer structure;
+ Spectrum spectrum;
+ Assignment assignment;
+ DataSet dataSet;
+ List spectraProperties1D;
+ String[] split;
+ String spectrumIndexInRecord;
+ List explicitHydrogenIndices;
+ int[] temp;
+
+ while (iterator.hasNext()) {
+ structure = iterator.next();
+ AtomContainerManipulator.percieveAtomTypesAndConfigureAtoms(structure);
+ explicitHydrogenIndices = casekit.nmr.utils.Utils.getExplicitHydrogenIndices(structure);
+ Collections.sort(explicitHydrogenIndices);
+ dataSet = Utils.atomContainerToDataSet(structure);
+
+ for (final String nucleus : nuclei) {
+ spectraProperties1D = getSpectraProperties1D(structure, nucleus);
+ for (final String spectrumProperty1D : spectraProperties1D) {
+ split = spectrumProperty1D.split("\\s");
+ spectrumIndexInRecord = split[split.length
+ - 1];
+
+ // skip molecules which do not contain any of requested spectrum information
+ spectrum = NMRShiftDBSpectrumToSpectrum(structure.getProperty(spectrumProperty1D), nucleus);
+ // if no spectrum could be built or the number of signals in spectrum is different than the atom number in molecule
+ if ((spectrum
+ == null)
+ || casekit.nmr.utils.Utils.getDifferenceSpectrumSizeAndMolecularFormulaCount(spectrum,
+ Utils.getMolecularFormulaFromString(
+ dataSet.getMeta()
+ .get("mf")),
+ 0)
+ != 0) {
+ continue;
+ }
+ if (structure.getProperty("Solvent")
+ != null) {
+ spectrum.addMetaInfo("solvent",
+ getSolvent(structure.getProperty("Solvent"), spectrumIndexInRecord));
+ }
+ if (structure.getProperty("Field Strength [MHz]")
+ != null) {
+ for (final String fieldStrength : structure.getProperty("Field Strength [MHz]")
+ .toString()
+ .split("\\s")) {
+ if (fieldStrength.startsWith(spectrumIndexInRecord
+ + ":")) {
+ try {
+ spectrum.addMetaInfo("spectrometerFrequency", fieldStrength.split(
+ spectrumIndexInRecord
+ + ":")[1]);
+ } catch (final NumberFormatException e) {
+ // e.printStackTrace();
+ }
+ break;
+ }
+ }
+ }
+
+ assignment = NMRShiftDBSpectrumToAssignment(structure.getProperty(spectrumProperty1D), nucleus);
+ if (assignment
+ != null
+ && !explicitHydrogenIndices.isEmpty()) {
+ int hCount;
+ for (int i = 0; i
+ < assignment.getSize(); i++) {
+ for (int k = 0; k
+ < assignment.getAssignment(0, i).length; k++) {
+ hCount = 0;
+ for (int j = 0; j
+ < explicitHydrogenIndices.size(); j++) {
+ if (explicitHydrogenIndices.get(j)
+ >= assignment.getAssignment(0, i, k)) {
+ break;
+ }
+ hCount++;
+ }
+ temp = assignment.getAssignment(0, i);
+ temp[k] = assignment.getAssignment(0, i, k)
+ - hCount;
+ assignment.setAssignment(0, i, temp);
+ }
+ }
+ }
+ dataSet.setSpectrum(new SpectrumCompact(spectrum));
+ dataSet.setAssignment(assignment);
+
+ dataSets.add(dataSet.buildClone());
+ }
+ }
+ }
+
+ return dataSets;
+ }
+
+ // /**
+ // * Returns a hashmap containing combined keys (by "_") of solvents
+ // * and lists of calculated deviations between all given spectra for a
+ // * nucleus in molecule record as values.
+ // * Here, only molecule records in NMRShiftDB file are considered which have
+ // * at least two different spectra for same nucleus.
+ // * Example: "Spectrum 13C 0", "Spectrum 13C 1" will be used for given
+ // * nucleus 13C.
+ // *
+ // * @param pathToNMRShiftDB
+ // * @param nucleus
+ // *
+ // * @return
+ // *
+ // * @throws FileNotFoundException
+ // * @throws CDKException
+ // */
+ // public static HashMap> getSolventDeviations(final String pathToNMRShiftDB, final String nucleus) throws FileNotFoundException, CDKException {
+ // int signalCount;
+ // Spectrum spectrum;
+ // Assignment assignment;
+ // final ArrayList> spectraSets = getSpectraFromNMRShiftDB(pathToNMRShiftDB, nucleus);
+ // HashMap> shiftsPerAtom;
+ // HashMap> solventsPerAtom;
+ // ArrayList solvents;
+ // String[] solventsToSort;
+ //
+ // final HashMap> deviations = new HashMap<>();
+ // String combiKey;
+ //
+ // for (final ArrayList spectraSetInRecord : spectraSets) {
+ // shiftsPerAtom = new HashMap<>();
+ // solventsPerAtom = new HashMap<>();
+ // signalCount = -1;
+ // for (final Object[] spectrumAndAssignment : spectraSetInRecord) {
+ // spectrum = (Spectrum) spectrumAndAssignment[0];
+ // assignment = (Assignment) spectrumAndAssignment[1];
+ // if (signalCount == -1) {
+ // signalCount = spectrum.getSignalCount();
+ // } else if (signalCount != spectrum.getSignalCount()) {
+ // continue;
+ // }
+ // for (final int atomIndex : assignment.getAssignments(0)) {
+ // if (!shiftsPerAtom.containsKey(atomIndex)) {
+ // shiftsPerAtom.put(atomIndex, new ArrayList<>());
+ // solventsPerAtom.put(atomIndex, new ArrayList<>());
+ // }
+ // shiftsPerAtom.get(atomIndex).add(spectrum.getSignal(assignment.getIndex(0, atomIndex)).getShift(0));
+ // solventsPerAtom.get(atomIndex).add(spectrum.getSolvent());
+ // }
+ // }
+ // if (shiftsPerAtom.isEmpty() || (shiftsPerAtom.get(Collections.min(shiftsPerAtom.keySet())).size() < 2)) {
+ // continue;
+ // }
+ // solvents = new ArrayList<>(solventsPerAtom.get(Collections.min(solventsPerAtom.keySet())));
+ // // if(Collections.frequency(solvents, "Unreported") + Collections.frequency(solvents, "Unknown") > solvents.size() - 2){
+ // // continue;
+ // // }
+ //
+ // for (final int atomIndex : shiftsPerAtom.keySet()) {
+ // for (int s1 = 0; s1 < solvents.size(); s1++) {
+ // // if(solvents.get(s1).equals("Unreported") || solvents.get(s1).equals("Unknown")){
+ // // continue;
+ // // }
+ // for (int s2 = s1 + 1; s2 < solvents.size(); s2++) {
+ // // if (solvents.get(s2).equals("Unreported") || solvents.get(s2).equals("Unknown")) {
+ // // continue;
+ // // }
+ // solventsToSort = new String[2];
+ // solventsToSort[0] = solvents.get(s1);
+ // solventsToSort[1] = solvents.get(s2);
+ // Arrays.sort(solventsToSort);
+ // combiKey = solventsToSort[0] + "_" + solventsToSort[1];
+ // if (!deviations.containsKey(combiKey)) {
+ // deviations.put(combiKey, new ArrayList<>());
+ // }
+ // deviations.get(combiKey).add(Math.abs(shiftsPerAtom.get(atomIndex).get(s1) - shiftsPerAtom.get(atomIndex).get(s2)));
+ // }
+ // }
+ // }
+ // }
+ //
+ // return deviations;
+ // }
+ //
+ // /**
+ // * @param pathToDB
+ // *
+ // * @return
+ // *
+ // * @throws FileNotFoundException
+ // * @deprecated
+ // */
+ // public static Set getAtomTypesInDB(final String pathToDB) throws FileNotFoundException {
+ // final HashSet atomTypes = new HashSet<>();
+ // final IteratingSDFReader iterator = new IteratingSDFReader(new FileReader(pathToDB), SilentChemObjectBuilder.getInstance());
+ // while (iterator.hasNext()) {
+ // atomTypes.addAll(HOSECodeUtilities.getAtomTypesInAtomContainer(iterator.next()));
+ // }
+ //
+ // return atomTypes;
+ // }
+
+ /**
+ * Creates a two dimensional array of a given NMRShiftDB casekit.nmr entry
+ * with all signal shift values, intensities, multiplicities and atom indices.
+ *
+ * @param NMRShiftDBSpectrum
+ *
+ * @return two dimensional array:
+ * 1. dimension: signal index (row);
+ * 2. dimension: signal shift value (column 1), signal intensity (column 2),
+ * signal multiplicity (column 3), atom index in structure (column 4)
+ */
+ public static String[][] parseNMRShiftDBSpectrum(final String NMRShiftDBSpectrum) {
+ if (NMRShiftDBSpectrum.trim()
+ .isEmpty()) {
+ return new String[][]{};
+ }
+ String[] signalSplit;
+ final String[] shiftsSplit = NMRShiftDBSpectrum.split("\\|");
+ final String[][] values = new String[shiftsSplit.length][4];
+ for (int i = 0; i
+ < shiftsSplit.length; i++) {
+ signalSplit = shiftsSplit[i].split(";");
+ values[i][0] = signalSplit[0]; // shift value
+ values[i][1] = signalSplit[1].toLowerCase()
+ .split("[a-z]")[0]; // intensity
+ values[i][2] = signalSplit[1].split("\\d+\\.\\d+").length
+ > 0
+ ? signalSplit[1].split("\\d+\\.\\d+")[1].toLowerCase()
+ : ""; // multiplicity
+ values[i][3] = signalSplit[2]; // atom index
+ }
+
+ return values;
+ }
+
+ public static Spectrum NMRShiftDBSpectrumToSpectrum(final String NMRShiftDBSpectrum, final String nucleus) {
+ if ((NMRShiftDBSpectrum
+ == null)
+ || NMRShiftDBSpectrum.trim()
+ .isEmpty()) {
+ return null;
+ }
+ final String[][] spectrumStringArray = parseNMRShiftDBSpectrum(NMRShiftDBSpectrum);
+ final Spectrum spectrum = new Spectrum();
+ spectrum.setNuclei(new String[]{nucleus});
+ spectrum.setSignals(new ArrayList<>());
+ String multiplicity;
+ Double shift, intensity;
+ try {
+ for (int i = 0; i
+ < spectrumStringArray.length; i++) {
+ shift = Double.parseDouble(spectrumStringArray[i][0]);
+ intensity = Double.parseDouble(spectrumStringArray[i][1]);
+ multiplicity = spectrumStringArray[i][2].trim()
+ .isEmpty()
+ ? null
+ : spectrumStringArray[i][2].trim()
+ .toLowerCase();
+ spectrum.addSignal(
+ new Signal(new String[]{nucleus}, new Double[]{shift}, multiplicity, "signal", intensity, 1, 0,
+ null, null));
+ }
+ } catch (final Exception e) {
+ return null;
+ }
+
+ return spectrum;
+ }
+
+ public static Assignment NMRShiftDBSpectrumToAssignment(final String NMRShiftDBSpectrum, final String nucleus) {
+ if ((NMRShiftDBSpectrum
+ == null)
+ || NMRShiftDBSpectrum.trim()
+ .isEmpty()) {
+ return null;
+ }
+ final String[][] NMRShiftDBSpectrumStringArray = parseNMRShiftDBSpectrum(NMRShiftDBSpectrum);
+ final Spectrum spectrum = NMRShiftDBSpectrumToSpectrum(NMRShiftDBSpectrum, nucleus);
+ final Assignment assignment = new Assignment();
+ assignment.setNuclei(spectrum.getNuclei());
+ assignment.initAssignments(spectrum.getSignalCount());
+ int signalIndex;
+ String multiplicity;
+ List closestSignalList;
+ for (int i = 0; i
+ < NMRShiftDBSpectrumStringArray.length; i++) {
+ // just to be sure that we take the right signal if equivalences are present
+ closestSignalList = spectrum.pickByClosestShift(Double.parseDouble(NMRShiftDBSpectrumStringArray[i][0]), 0,
+ 0.0);
+ multiplicity = NMRShiftDBSpectrumStringArray[i][2].trim()
+ .isEmpty()
+ ? null
+ : NMRShiftDBSpectrumStringArray[i][2].trim()
+ .toLowerCase();
+ closestSignalList.retainAll(spectrum.pickByMultiplicity(multiplicity));
+ signalIndex = closestSignalList.get(0);
+
+ assignment.addAssignmentEquivalence(0, signalIndex, Integer.parseInt(NMRShiftDBSpectrumStringArray[i][3]));
+ }
+
+ return assignment;
+ }
+}
diff --git a/src/casekit/nmr/dbservice/Utils.java b/src/casekit/nmr/dbservice/Utils.java
new file mode 100644
index 0000000..4fb1bf2
--- /dev/null
+++ b/src/casekit/nmr/dbservice/Utils.java
@@ -0,0 +1,53 @@
+package casekit.nmr.dbservice;
+
+import org.openscience.cdk.exception.CDKException;
+import org.openscience.cdk.io.SDFWriter;
+import org.openscience.cdk.io.iterator.IteratingSDFReader;
+import org.openscience.cdk.silent.SilentChemObjectBuilder;
+
+import java.io.BufferedWriter;
+import java.io.FileReader;
+import java.io.FileWriter;
+import java.io.IOException;
+
+public class Utils {
+
+ public static int splitSDFile(final String pathToSDFile, final int maxMolPerFile) throws IOException, CDKException {
+ final IteratingSDFReader iterator = new IteratingSDFReader(new FileReader(pathToSDFile),
+ SilentChemObjectBuilder.getInstance());
+ final String fileEnding = pathToSDFile.split("\\.")[1];
+ BufferedWriter bufferedWriter;
+ SDFWriter sdfWriter;
+ int counter = 0;
+ int part = 1;
+ bufferedWriter = new BufferedWriter(new FileWriter(pathToSDFile.split("\\.")[0]
+ + "_"
+ + part
+ + "."
+ + fileEnding));
+ sdfWriter = new SDFWriter(bufferedWriter);
+ while (iterator.hasNext()) {
+ if (counter
+ >= maxMolPerFile) {
+ sdfWriter.close();
+
+ part++;
+ bufferedWriter = new BufferedWriter(new FileWriter(pathToSDFile.split("\\.")[0]
+ + "_"
+ + part
+ + "."
+ + fileEnding));
+ sdfWriter = new SDFWriter(bufferedWriter);
+ sdfWriter.write(iterator.next());
+ counter = 1;
+ } else {
+ sdfWriter.write(iterator.next());
+ }
+ counter++;
+ }
+ sdfWriter.close();
+
+ return part
+ - 1;
+ }
+}
diff --git a/src/casekit/nmr/elucidation/Constants.java b/src/casekit/nmr/elucidation/Constants.java
new file mode 100644
index 0000000..02ab23b
--- /dev/null
+++ b/src/casekit/nmr/elucidation/Constants.java
@@ -0,0 +1,96 @@
+package casekit.nmr.elucidation;
+
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+public class Constants {
+
+ // valid strings from LSD webpage: C N N5 O S S4 S6 F Cl Br I P P5 Si B X
+ public static final Map nucleiMap = createNucleiMap();
+ public static final Map defaultHybridizationMap = createDefaultHybridizationMap();
+ public static final Map defaultProtonsCountPerValencyMap = createDefaultProtonsCountPerValencyMap();
+ public static final Map defaultAtomLabelMap = createDefaultAtomLabelMap();
+ public static final Map hybridizationConversionMap = createHybridizationConversionMap();
+
+ private static Map createNucleiMap() {
+ final Map nuclei = new HashMap<>();
+ nuclei.put("C", "13C");
+ nuclei.put("N", "15N");
+ nuclei.put("H", "1H");
+ nuclei.put("S", "33S");
+ nuclei.put("F", "19F");
+ nuclei.put("P", "31P");
+ nuclei.put("Si", "29Si");
+
+ return Collections.unmodifiableMap(nuclei);
+ }
+
+ private static Map createDefaultHybridizationMap() {
+ final Map defaultHybridization = new HashMap<>();
+ defaultHybridization.put("C", new int[]{1, 2, 3});
+ defaultHybridization.put("N", new int[]{1, 2, 3});
+ defaultHybridization.put("S", new int[]{1, 2, 3});
+ defaultHybridization.put("O", new int[]{2, 3});
+ defaultHybridization.put("I", new int[]{3});
+ defaultHybridization.put("F", new int[]{3});
+ defaultHybridization.put("Cl", new int[]{3});
+ defaultHybridization.put("Br", new int[]{3});
+ defaultHybridization.put("P", new int[]{1, 2, 3});
+ defaultHybridization.put("Si", new int[]{1, 2, 3});
+
+ return Collections.unmodifiableMap(defaultHybridization);
+ }
+
+ private static Map createDefaultProtonsCountPerValencyMap() {
+ final Map defaultProtonsCountPerValency = new HashMap<>();
+ defaultProtonsCountPerValency.put("C", new int[]{0, 1, 2, 3});
+ defaultProtonsCountPerValency.put("N", new int[]{0, 1, 2});
+ defaultProtonsCountPerValency.put("N5", new int[]{0, 1, 2, 3});
+ defaultProtonsCountPerValency.put("N35", new int[]{0, 1, 2, 3});
+ defaultProtonsCountPerValency.put("S", new int[]{0, 1});
+ defaultProtonsCountPerValency.put("S4", new int[]{0, 1, 2, 3});
+ defaultProtonsCountPerValency.put("S6", new int[]{0, 1, 2, 3});
+ defaultProtonsCountPerValency.put("S246", new int[]{0, 1, 2, 3});
+ defaultProtonsCountPerValency.put("O", new int[]{0, 1});
+ defaultProtonsCountPerValency.put("I", new int[]{0});
+ defaultProtonsCountPerValency.put("F", new int[]{0});
+ defaultProtonsCountPerValency.put("Cl", new int[]{0});
+ defaultProtonsCountPerValency.put("Br", new int[]{0});
+ defaultProtonsCountPerValency.put("P", new int[]{0, 1, 2});
+ defaultProtonsCountPerValency.put("P5", new int[]{0, 1, 2, 3});
+ defaultProtonsCountPerValency.put("P35", new int[]{0, 1, 2, 3});
+ defaultProtonsCountPerValency.put("Si", new int[]{0, 1, 2, 3});
+
+ return defaultProtonsCountPerValency;
+ }
+
+ private static Map createDefaultAtomLabelMap() {
+ final Map defaultAtomLabel = new HashMap<>();
+ defaultAtomLabel.put("C", "C");
+ defaultAtomLabel.put("N", "N35");
+ defaultAtomLabel.put("O", "O");
+ defaultAtomLabel.put("S", "S246");
+ defaultAtomLabel.put("I", "I");
+ defaultAtomLabel.put("F", "F");
+ defaultAtomLabel.put("Cl", "Cl");
+ defaultAtomLabel.put("Br", "Br");
+ defaultAtomLabel.put("P", "P35");
+ defaultAtomLabel.put("Si", "Si");
+
+ return Collections.unmodifiableMap(defaultAtomLabel);
+ }
+
+ private static Map createHybridizationConversionMap() {
+ // @TODO access this information from MongoDB and store it instead of hard coding it
+ // possible command in MongoDB: db.hybridizations.aggregate([{$match: {nucleus: "15N"}}, {$group: {_id: null, set: {$addToSet: "$hybridization"}}}])
+ // nucleus -> hybridization string -> number
+ final Map hybridizationConversionMap = new HashMap<>();
+ hybridizationConversionMap.put("PLANAR3", 3);
+ hybridizationConversionMap.put("SP3", 3);
+ hybridizationConversionMap.put("SP2", 2);
+ hybridizationConversionMap.put("SP1", 1);
+
+ return Collections.unmodifiableMap(hybridizationConversionMap);
+ }
+}
diff --git a/src/casekit/nmr/elucidation/Utilities.java b/src/casekit/nmr/elucidation/Utilities.java
new file mode 100644
index 0000000..2528d03
--- /dev/null
+++ b/src/casekit/nmr/elucidation/Utilities.java
@@ -0,0 +1,1114 @@
+package casekit.nmr.elucidation;
+
+import casekit.nmr.elucidation.model.Detections;
+import casekit.nmr.elucidation.model.Grouping;
+import casekit.nmr.elucidation.model.MolecularConnectivity;
+import casekit.nmr.model.Signal;
+import casekit.nmr.model.nmrium.Correlation;
+import casekit.nmr.model.nmrium.Link;
+import casekit.nmr.utils.Utils;
+
+import java.util.*;
+import java.util.stream.Collectors;
+
+public class Utilities {
+
+ public static void reduceDefaultHybridizationsAndProtonCountsOfHeteroAtoms(final List correlationList,
+ final Map>>> detectedConnectivities,
+ final Map> detectedHybridizations) {
+ if (detectedConnectivities
+ == null
+ || detectedConnectivities.isEmpty()) {
+ return;
+ }
+ final Map> allowedNeighborAtomHybridizations = buildAllowedNeighborAtomHybridizations(
+ correlationList, detectedConnectivities);
+ final Map> allowedNeighborAtomProtonCounts = buildAllowedNeighborAtomProtonCounts(
+ correlationList, detectedConnectivities);
+ // hetero atoms can bond to carbons only, due to that we can use further connectivity information
+ // do not allow bond between carbon and hetero atoms in certain hybridization states and proton counts
+ Correlation correlation;
+ for (int i = 0; i
+ < correlationList.size(); i++) {
+ correlation = correlationList.get(i);
+ // ignore C and H atoms
+ if (correlation.getAtomType()
+ .equals("C")
+ || correlation.getAtomType()
+ .equals("H")) {
+ continue;
+ }
+ final Set hybridizationsToAdd = allowedNeighborAtomHybridizations.containsKey(
+ correlation.getAtomType())
+ ? allowedNeighborAtomHybridizations.get(correlation.getAtomType())
+ : Arrays.stream(Constants.defaultHybridizationMap.get(
+ correlation.getAtomType()))
+ .boxed()
+ .collect(Collectors.toSet());
+ final Set protonCountsToAdd = allowedNeighborAtomProtonCounts.containsKey(
+ correlation.getAtomType())
+ ? allowedNeighborAtomProtonCounts.get(correlation.getAtomType())
+ : Arrays.stream(Constants.defaultProtonsCountPerValencyMap.get(
+ correlation.getAtomType()))
+ .boxed()
+ .collect(Collectors.toSet());
+ // but only if we have seen the hetero atom type in connectivity statistics
+ // and hybridization states or protons count was not set beforehand
+ if (correlation.getHybridization()
+ .isEmpty()) {
+ correlation.getHybridization()
+ .addAll(hybridizationsToAdd);
+ } else if (correlation.getEdited()
+ != null
+ && correlation.getEdited()
+ .containsKey("hybridization")
+ && !correlation.getEdited()
+ .get("hybridization")
+ && allowedNeighborAtomHybridizations.containsKey(correlation.getAtomType())) {
+ correlation.getHybridization()
+ .retainAll(hybridizationsToAdd);
+ }
+ if (correlation.getProtonsCount()
+ .isEmpty()) {
+ correlation.getProtonsCount()
+ .addAll(protonCountsToAdd);
+ } else if (correlation.getEdited()
+ != null
+ && correlation.getEdited()
+ .containsKey("protonsCount")
+ && !correlation.getEdited()
+ .get("protonsCount")) {
+ correlation.getProtonsCount()
+ .retainAll(protonCountsToAdd);
+ }
+ detectedHybridizations.putIfAbsent(i, new ArrayList<>());
+ detectedHybridizations.get(i)
+ .addAll(correlation.getHybridization());
+ }
+ }
+
+ public static Map> buildAllowedNeighborAtomHybridizations(
+ final List correlationList,
+ final Map>>> detectedConnectivities) {
+ final Map> allowedHeteroAtomHybridizations = new HashMap<>();
+ for (final Map.Entry>>> correlationEntry : detectedConnectivities.entrySet()) {
+ if (!correlationList.get(correlationEntry.getKey())
+ .getAtomType()
+ .equals("C")
+ && !correlationList.get(correlationEntry.getKey())
+ .getAtomType()
+ .equals("H")) {
+ continue;
+ }
+ for (final Map.Entry>> neighborAtomTypeEntry : correlationEntry.getValue()
+ .entrySet()) {
+ allowedHeteroAtomHybridizations.putIfAbsent(neighborAtomTypeEntry.getKey(), new HashSet<>());
+ allowedHeteroAtomHybridizations.get(neighborAtomTypeEntry.getKey())
+ .addAll(neighborAtomTypeEntry.getValue()
+ .keySet());
+ }
+ }
+
+ return allowedHeteroAtomHybridizations;
+ }
+
+ public static Map> buildAllowedNeighborAtomProtonCounts(
+ final List correlationList,
+ final Map>>> detectedConnectivities) {
+ final Map> allowedHeteroAtomProtonCounts = new HashMap<>();
+ for (final Map.Entry>>> correlationEntry : detectedConnectivities.entrySet()) {
+ if (!correlationList.get(correlationEntry.getKey())
+ .getAtomType()
+ .equals("C")
+ && !correlationList.get(correlationEntry.getKey())
+ .getAtomType()
+ .equals("H")) {
+ continue;
+ }
+ for (final Map.Entry>> neighborAtomTypeEntry : correlationEntry.getValue()
+ .entrySet()) {
+ allowedHeteroAtomProtonCounts.putIfAbsent(neighborAtomTypeEntry.getKey(), new HashSet<>());
+ for (final Map.Entry> neighborHybridizationEntry : neighborAtomTypeEntry.getValue()
+ .entrySet()) {
+ allowedHeteroAtomProtonCounts.get(neighborAtomTypeEntry.getKey())
+ .addAll(neighborHybridizationEntry.getValue());
+ }
+ }
+ }
+
+ return allowedHeteroAtomProtonCounts;
+ }
+
+ public static Map>> buildForbiddenNeighbors(
+ final Map>> connectivities, final Set possibleNeighborAtomTypes) {
+
+ // define forbidden neighbors (for carbons only)
+ // or put just an empty map which means the whole element is forbidden
+ final Map>> forbiddenNeighbors = new HashMap<>();
+ for (final String possibleNeighborAtomType : possibleNeighborAtomTypes) {
+ if (possibleNeighborAtomType.equals("H")) {
+ continue;
+ }
+ forbiddenNeighbors.put(possibleNeighborAtomType, new HashMap<>());
+ if (connectivities.containsKey(possibleNeighborAtomType)) {
+ for (final int defaultHybridization : Arrays.stream(
+ Constants.defaultHybridizationMap.get(possibleNeighborAtomType))
+ .boxed()
+ .collect(Collectors.toList())) {
+ forbiddenNeighbors.get(possibleNeighborAtomType)
+ .put(defaultHybridization, Arrays.stream(
+ Constants.defaultProtonsCountPerValencyMap.get(possibleNeighborAtomType))
+ .boxed()
+ .collect(Collectors.toSet()));
+ }
+ for (final int possibleNeighborHybridization : connectivities.get(possibleNeighborAtomType)
+ .keySet()) {
+ // remove found protons count per hybridzations from list of forbidden ones
+ for (final int forbiddenNeighborHybridization : new HashSet<>(
+ forbiddenNeighbors.get(possibleNeighborAtomType)
+ .keySet())) {
+ forbiddenNeighbors.get(possibleNeighborAtomType)
+ .get(forbiddenNeighborHybridization)
+ .removeAll(connectivities.get(possibleNeighborAtomType)
+ .get(possibleNeighborHybridization));
+ if (forbiddenNeighbors.get(possibleNeighborAtomType)
+ .get(forbiddenNeighborHybridization)
+ .isEmpty()) {
+ forbiddenNeighbors.get(possibleNeighborAtomType)
+ .remove(forbiddenNeighborHybridization);
+ }
+ }
+ if (forbiddenNeighbors.get(possibleNeighborAtomType)
+ .isEmpty()) {
+ forbiddenNeighbors.remove(possibleNeighborAtomType);
+ break;
+ }
+ }
+ }
+ }
+
+ return forbiddenNeighbors;
+ }
+
+
+ public static Map> buildFixedNeighborsByINADEQUATE(final List correlationList) {
+ final Map> fixedNeighbors = new HashMap<>();
+ final Set uniqueSet = new HashSet<>();
+ Correlation correlation;
+ for (int i = 0; i
+ < correlationList.size(); i++) {
+ correlation = correlationList.get(i);
+ // @TODO for now use INADEQUATE information of atoms without equivalences only
+ if (correlation.getEquivalence()
+ != 1) {
+ continue;
+ }
+ for (final Link link : correlation.getLink()) {
+ if (link.getExperimentType()
+ .equals("inadequate")) {
+ for (final int matchIndex : link.getMatch()) {
+ // insert BOND pair once only and not if equivalences exist
+ if (!uniqueSet.contains(i
+ + " "
+ + matchIndex)
+ && correlationList.get(matchIndex)
+ .getEquivalence()
+ == 1) {
+ fixedNeighbors.putIfAbsent(i, new HashSet<>());
+ fixedNeighbors.get(i)
+ .add(matchIndex);
+ uniqueSet.add(i
+ + " "
+ + matchIndex);
+ uniqueSet.add(matchIndex
+ + " "
+ + i);
+ }
+ }
+ }
+ }
+ }
+
+ return fixedNeighbors;
+ }
+
+ private static boolean hasMatch(final Correlation correlation1, final Correlation correlation2,
+ final double tolerance) {
+ final Signal signal1 = Utils.extractFirstSignalFromCorrelation(correlation1);
+ final Signal signal2 = Utils.extractFirstSignalFromCorrelation(correlation2);
+ if (signal1
+ == null
+ || signal2
+ == null) {
+ return false;
+ }
+ int dim1 = -1;
+ int dim2 = -1;
+ String atomType;
+ for (int i = 0; i
+ < signal1.getNuclei().length; i++) {
+ atomType = Utils.getAtomTypeFromNucleus(signal1.getNuclei()[i]);
+ if (atomType.equals(correlation1.getAtomType())) {
+ dim1 = i;
+ break;
+ }
+ }
+ for (int i = 0; i
+ < signal2.getNuclei().length; i++) {
+ atomType = Utils.getAtomTypeFromNucleus(signal2.getNuclei()[i]);
+ if (atomType.equals(correlation2.getAtomType())) {
+ dim2 = i;
+ break;
+ }
+ }
+ if (dim1
+ == -1
+ || dim2
+ == -1) {
+ return false;
+ }
+
+ final double shift1 = signal1.getShift(dim1);
+ final double shift2 = signal2.getShift(dim2);
+
+ return Math.abs(shift1
+ - shift2)
+ <= tolerance;
+
+ }
+
+ private static Map>> findGroups(final List correlationList,
+ final Map tolerances) {
+ // cluster group index -> list of correlation index pair
+ final Map>> groups = new HashMap<>();
+ int groupIndex = 0;
+ final Set inserted = new HashSet<>();
+ int foundGroupIndex;
+ for (int i = 0; i
+ < correlationList.size(); i++) {
+ final Correlation correlation = correlationList.get(i);
+ if (inserted.contains(i)
+ || correlation.isPseudo()) {
+ continue;
+ }
+ groups.putIfAbsent(correlation.getAtomType(), new HashMap<>());
+ // if we have a match somewhere then add the correlation index into to group
+ // if not then create a new group
+ foundGroupIndex = -1;
+ for (final Map.Entry> groupEntry : groups.get(correlation.getAtomType())
+ .entrySet()) {
+ if (groupEntry.getValue()
+ .stream()
+ .anyMatch(correlationIndex -> hasMatch(correlation, correlationList.get(correlationIndex),
+ tolerances.get(correlation.getAtomType())))) {
+ foundGroupIndex = groupEntry.getKey();
+ break;
+ }
+ }
+ if (foundGroupIndex
+ != -1) {
+ groups.get(correlation.getAtomType())
+ .get(foundGroupIndex)
+ .add(i);
+ inserted.add(i);
+ } else {
+ groups.get(correlation.getAtomType())
+ .put(groupIndex, new ArrayList<>());
+ groups.get(correlation.getAtomType())
+ .get(groupIndex)
+ .add(i);
+ inserted.add(i);
+ groupIndex++;
+ }
+ }
+
+ return groups;
+ }
+
+ private static Map> transformGroups(
+ final Map>> groups) {
+ final Map> transformedGroups = new HashMap<>();
+ for (final Map.Entry>> atomTypeEntry : groups.entrySet()) {
+ transformedGroups.put(atomTypeEntry.getKey(), new HashMap<>());
+ for (final Map.Entry> groupEntry : atomTypeEntry.getValue()
+ .entrySet()) {
+ for (final int correlationIndex : groupEntry.getValue()) {
+ transformedGroups.get(atomTypeEntry.getKey())
+ .put(correlationIndex, groupEntry.getKey());
+ }
+ }
+ }
+
+ return transformedGroups;
+ }
+
+ public static Grouping buildGroups(final List correlationList, final Map tolerances) {
+ final Map>> groups = findGroups(correlationList, tolerances);
+
+ return new Grouping(tolerances, groups, transformGroups(groups));
+ }
+
+ private static List getProtonCounts(final List correlationList, final int index) {
+ final Correlation correlation = correlationList.get(index);
+ if (correlation.getProtonsCount()
+ != null
+ && !correlation.getProtonsCount()
+ .isEmpty()) {
+ // if protonCounts is already given
+ return correlation.getProtonsCount();
+ }
+ final List protonCounts = new ArrayList<>();
+ final int[] defaultProtonCounts = Constants.defaultProtonsCountPerValencyMap.get(
+ Constants.defaultAtomLabelMap.get(correlation.getAtomType()));
+ for (final int defaultProtonCount : defaultProtonCounts) {
+ protonCounts.add(defaultProtonCount);
+ }
+
+ return protonCounts;
+ }
+
+ private static List getHybridizations(final List correlationList, final int index,
+ final Map> detectedHybridizations) {
+ final Correlation correlation = correlationList.get(index);
+ final Set hybridizations = new HashSet<>(correlation.getHybridization()
+ != null
+ ? correlation.getHybridization()
+ : new ArrayList<>());
+ if (detectedHybridizations.containsKey(index)) {
+ hybridizations.addAll(detectedHybridizations.get(index));
+ }
+ if (hybridizations.isEmpty()
+ && correlation.getAtomType()
+ .equals("C")
+ && correlation.getProtonsCount()
+ .size()
+ == 1) {
+ if (correlation.getProtonsCount()
+ .get(0)
+ == 2) {
+ // a carbon with two protons can only be SP2 or SP3
+ hybridizations.add(2);
+ hybridizations.add(3);
+ } else if (correlation.getProtonsCount()
+ .get(0)
+ == 3) {
+ // a carbon with three protons can only be SP3
+ hybridizations.add(3);
+ }
+ }
+ if (hybridizations.isEmpty()) {
+ for (int i = 0; i
+ < Constants.defaultHybridizationMap.get(correlation.getAtomType()).length; i++) {
+ hybridizations.add(Constants.defaultHybridizationMap.get(correlation.getAtomType())[i]);
+ }
+ }
+
+ return new ArrayList<>(hybridizations);
+ }
+
+ public static Map buildIndicesMap(final List molecularConnectivityList) {
+ // index in correlation data -> [indices in PyLSD file...]
+ final Map indicesMap = new HashMap<>();
+ // init element indices within correlations with same order as in correlation data input
+ int heavyAtomIndexInPyLSDFile = 1;
+ int protonIndexInPyLSDFile = 1;
+ int protonsToInsert;
+ Integer[] arrayToInsert;
+ for (final MolecularConnectivity molecularConnectivity : molecularConnectivityList) {
+ // set entry for each correlation with consideration of equivalences
+ if (!molecularConnectivity.getAtomType()
+ .equals("H")) {
+ if (molecularConnectivity.getHsqc()
+ != null) {
+ // insert for protons
+ protonsToInsert = molecularConnectivity.getEquivalence();
+ arrayToInsert = new Integer[protonsToInsert];
+ for (int j = 0; j
+ < arrayToInsert.length; j++) {
+ arrayToInsert[j] = protonIndexInPyLSDFile;
+ protonIndexInPyLSDFile++;
+ }
+ for (final int attachedProtonIndex : molecularConnectivity.getHsqc()) {
+ indicesMap.put(attachedProtonIndex, arrayToInsert);
+ }
+ }
+ // insert for heavy atom itself
+ indicesMap.put(molecularConnectivity.getIndex(), new Integer[molecularConnectivity.getEquivalence()]);
+ for (int j = 0; j
+ < molecularConnectivity.getEquivalence(); j++) {
+ indicesMap.get(molecularConnectivity.getIndex())[j] = heavyAtomIndexInPyLSDFile;
+ heavyAtomIndexInPyLSDFile++;
+ }
+ }
+ }
+
+ return indicesMap;
+ }
+
+ public static MolecularConnectivity findMolecularConnectivityByIndex(
+ final Map> molecularConnectivityMap, final String atomType,
+ final boolean exclude, final int index) {
+ for (final int correlationIndex : molecularConnectivityMap.keySet()) {
+ for (final MolecularConnectivity molecularConnectivity : molecularConnectivityMap.get(correlationIndex)) {
+ if (((exclude
+ && !molecularConnectivity.getAtomType()
+ .equals(atomType))
+ || (!exclude
+ && molecularConnectivity.getAtomType()
+ .equals(atomType)))
+ && molecularConnectivity.getIndex()
+ == index) {
+ return molecularConnectivity;
+ }
+ }
+ }
+
+ return null;
+ }
+
+ private static List buildPossibilities(final Map indicesMap,
+ final List molecularConnectivityList,
+ final int correlationIndex, final Grouping grouping,
+ final boolean useGrouping) {
+ final MolecularConnectivity molecularConnectivity = molecularConnectivityList.get(correlationIndex);
+ // add possible indices from grouping
+ final int groupIndex;
+ final Set possibilities = new HashSet<>();
+ if (useGrouping
+ && grouping.getTransformedGroups()
+ .containsKey(molecularConnectivity.getAtomType())
+ && grouping.getTransformedGroups()
+ .get(molecularConnectivity.getAtomType())
+ .containsKey(correlationIndex)) {
+ groupIndex = grouping.getTransformedGroups()
+ .get(molecularConnectivity.getAtomType())
+ .get(correlationIndex);
+ for (final int groupMemberIndex : grouping.getGroups()
+ .get(molecularConnectivity.getAtomType())
+ .get(groupIndex)) {
+
+ if (indicesMap.containsKey(groupMemberIndex)) {
+ // add equivalence indices of group members
+ possibilities.addAll(Arrays.asList(indicesMap.get(groupMemberIndex)));
+ }
+ }
+ } else {
+ // add for equivalences only
+ possibilities.addAll(Arrays.asList(indicesMap.get(correlationIndex)));
+ }
+
+ return new ArrayList<>(possibilities);
+ }
+
+ public static Map> buildMolecularConnectivityMap(
+ final List molecularConnectivityList, final Detections detections,
+ final Grouping grouping, final Map defaultBondDistances) {
+
+ final Map indicesMap = buildIndicesMap(molecularConnectivityList);
+ // correlation index -> [MolecularConnectivity]
+ final Map> molecularConnectivityMap = new HashMap<>();
+ MolecularConnectivity newMolecularConnectivity;
+
+ int index, correlationIndex, protonCorrelationIndex, protonIndex;
+ for (final MolecularConnectivity molecularConnectivity : molecularConnectivityList) {
+ correlationIndex = molecularConnectivity.getIndex();
+ // skip in case of non-linked proton which has no index in indices map
+ if (!indicesMap.containsKey(correlationIndex)) {
+ continue;
+ }
+ for (int k = 0; k
+ < indicesMap.get(correlationIndex).length; k++) {
+ index = indicesMap.get(correlationIndex)[k];
+ newMolecularConnectivity = new MolecularConnectivity();
+ newMolecularConnectivity.setIndex(index);
+ newMolecularConnectivity.setAtomType(molecularConnectivity.getAtomType());
+ newMolecularConnectivity.setEquivalence(1);
+ newMolecularConnectivity.setPseudo(molecularConnectivity.isPseudo());
+ newMolecularConnectivity.setProtonCounts(molecularConnectivity.getProtonCounts());
+ newMolecularConnectivity.setHybridizations(molecularConnectivity.getHybridizations());
+ newMolecularConnectivity.setSignal(molecularConnectivity.getSignal());
+
+ if (!molecularConnectivity.getAtomType()
+ .equals("H")
+ && molecularConnectivity.getHsqc()
+ != null
+ && k
+ < indicesMap.get(molecularConnectivity.getHsqc()
+ .get(0)).length) {
+ // using the first proton correlation index from HSQC list is enough because show will direct to same heavy atom index
+ protonIndex = indicesMap.get(molecularConnectivity.getHsqc()
+ .get(0))[k];
+ newMolecularConnectivity.setHsqc(new ArrayList<>());
+ newMolecularConnectivity.getHsqc()
+ .add(protonIndex);
+ }
+ if (!molecularConnectivity.getAtomType()
+ .equals("H")
+ && molecularConnectivity.getHmbc()
+ != null) {
+ for (final Map.Entry entryHMBC : molecularConnectivity.getHmbc()
+ .entrySet()) {
+ protonCorrelationIndex = entryHMBC.getKey();
+ if (indicesMap.containsKey(protonCorrelationIndex)) {
+ for (int l = 0; l
+ < indicesMap.get(protonCorrelationIndex).length; l++) {
+ protonIndex = indicesMap.get(protonCorrelationIndex)[l];
+ if (newMolecularConnectivity.getHmbc()
+ == null) {
+ newMolecularConnectivity.setHmbc(new HashMap<>());
+ }
+ newMolecularConnectivity.getHmbc()
+ .put(protonIndex, entryHMBC.getValue());
+ }
+ }
+ }
+ }
+ if (molecularConnectivity.getAtomType()
+ .equals("H")
+ && molecularConnectivity.getCosy()
+ != null) {
+ for (final Map.Entry entryCOSY : molecularConnectivity.getCosy()
+ .entrySet()) {
+ protonCorrelationIndex = entryCOSY.getKey();
+ if (indicesMap.containsKey(protonCorrelationIndex)
+ && k
+ < indicesMap.get(protonCorrelationIndex).length) {
+ protonIndex = indicesMap.get(protonCorrelationIndex)[k];
+ if (newMolecularConnectivity.getCosy()
+ == null) {
+ newMolecularConnectivity.setCosy(new HashMap<>());
+ }
+ newMolecularConnectivity.getCosy()
+ .put(protonIndex, entryCOSY.getValue());
+ }
+ }
+ }
+
+ molecularConnectivityMap.putIfAbsent(correlationIndex, new ArrayList<>());
+ molecularConnectivityMap.get(correlationIndex)
+ .add(newMolecularConnectivity);
+ }
+ // set detections
+ for (final MolecularConnectivity molecularConnectivityTemp : molecularConnectivityMap.get(
+ correlationIndex)) {
+ if (detections.getForbiddenNeighbors()
+ .containsKey(correlationIndex)) {
+ molecularConnectivityTemp.setForbiddenNeighbors(detections.getForbiddenNeighbors()
+ .get(correlationIndex));
+ }
+ if (detections.getSetNeighbors()
+ .containsKey(correlationIndex)) {
+ molecularConnectivityTemp.setSetNeighbors(detections.getSetNeighbors()
+ .get(correlationIndex));
+ }
+ molecularConnectivityTemp.setGroupMembers(
+ buildPossibilities(indicesMap, molecularConnectivityList, correlationIndex, grouping,
+ // !molecularConnectivity.getAtomType()
+ // .equals("H"))
+ true));
+ }
+ // fill in fixed neighbors
+ if (molecularConnectivity.getEquivalence()
+ == 1
+ && detections.getFixedNeighbors()
+ .containsKey(correlationIndex)) {
+ MolecularConnectivity molecularConnectivityTemp;
+ for (final int correlationIndex2 : detections.getFixedNeighbors()
+ .get(correlationIndex)) {
+ molecularConnectivityTemp = molecularConnectivityList.get(correlationIndex2);
+ // use fixed neighbor information of atoms without equivalence equals 1 only
+ if (molecularConnectivityTemp.getEquivalence()
+ > 1) {
+ continue;
+ }
+ index = indicesMap.get(correlationIndex)[0];
+ newMolecularConnectivity = findMolecularConnectivityByIndex(molecularConnectivityMap,
+ molecularConnectivity.getAtomType(),
+ false, index);
+ if (newMolecularConnectivity.getFixedNeighbors()
+ == null) {
+ newMolecularConnectivity.setFixedNeighbors(new ArrayList<>());
+ }
+ newMolecularConnectivity.getFixedNeighbors()
+ .add(indicesMap.get(correlationIndex2)[0]);
+ }
+ }
+ }
+
+ // filter out HMBC or COSY correlation to itself
+ for (final int correlationIndexTemp : molecularConnectivityMap.keySet()) {
+ for (final MolecularConnectivity molecularConnectivityTemp : molecularConnectivityMap.get(
+ correlationIndexTemp)) {
+ if (molecularConnectivityTemp.getHsqc()
+ != null) {
+ if (molecularConnectivityTemp.getHmbc()
+ != null) {
+ molecularConnectivityTemp.getHmbc()
+ .remove(molecularConnectivityTemp.getHsqc());
+ }
+ if (molecularConnectivityTemp.getCosy()
+ != null) {
+ molecularConnectivityTemp.getCosy()
+ .remove(molecularConnectivityTemp.getHsqc());
+ }
+ }
+ }
+ }
+
+ return molecularConnectivityMap;
+ }
+
+ public static List buildMolecularConnectivityList(final List correlationList,
+ final Detections detections,
+ final Grouping grouping,
+ final Map defaultBondDistances) {
+ final List molecularConnectivityList = new ArrayList<>();
+ Correlation correlation;
+ int groupIndex;
+ Map signal2DMap;
+ Map jMap;
+ Map pathLengthMap;
+ MolecularConnectivity molecularConnectivity;
+ for (int correlationIndex = 0; correlationIndex
+ < correlationList.size(); correlationIndex++) {
+ correlation = correlationList.get(correlationIndex);
+ molecularConnectivity = new MolecularConnectivity();
+ molecularConnectivity.setIndex(correlationIndex);
+ molecularConnectivity.setAtomType(correlation.getAtomType());
+ molecularConnectivity.setSignal(Utils.extractFirstSignalFromCorrelation(correlation));
+ molecularConnectivity.setEquivalence(correlation.getEquivalence());
+ molecularConnectivity.setPseudo(correlation.isPseudo());
+
+ if (!correlation.getAtomType()
+ .equals("H")) {
+ molecularConnectivity.setProtonCounts(getProtonCounts(correlationList, correlationIndex));
+ molecularConnectivity.setHybridizations(
+ getHybridizations(correlationList, correlationIndex, detections.getDetectedHybridizations()));
+ }
+ if (grouping.getGroups()
+ .containsKey(correlation.getAtomType())
+ && grouping.getTransformedGroups()
+ .get(correlation.getAtomType())
+ .containsKey(correlationIndex)) {
+ groupIndex = grouping.getTransformedGroups()
+ .get(correlation.getAtomType())
+ .get(correlationIndex);
+ molecularConnectivity.setGroupMembers(grouping.getGroups()
+ .get(correlation.getAtomType())
+ .get(groupIndex));
+ }
+ for (final Link link : correlation.getLink()) {
+ if (link.getExperimentType()
+ .equals("hsqc")
+ || link.getExperimentType()
+ .equals("hmqc")
+ && !correlation.getAtomType()
+ .equals("H")) {
+ if (molecularConnectivity.getHsqc()
+ == null) {
+ molecularConnectivity.setHsqc(new ArrayList<>());
+ }
+ for (final int matchIndex : link.getMatch()) {
+ molecularConnectivity.getHsqc()
+ .add(matchIndex);
+ }
+ } else if (link.getExperimentType()
+ .equals("hmbc")
+ || link.getExperimentType()
+ .equals("cosy")) {
+ if (link.getExperimentType()
+ .equals("hmbc")
+ && correlation.getAtomType()
+ .equals("H")) {
+ continue;
+ }
+ if (link.getExperimentType()
+ .equals("cosy")
+ && !correlation.getAtomType()
+ .equals("H")) {
+ continue;
+ }
+ // ignore H atoms without any attachment to a heavy atom
+ if (correlationList.get(correlationIndex)
+ .getAtomType()
+ .equals("H")
+ && correlationList.get(correlationIndex)
+ .getAttachment()
+ .keySet()
+ .isEmpty()) {
+ continue;
+ }
+ signal2DMap = (Map) link.getSignal();
+ if (signal2DMap
+ != null
+ && signal2DMap.containsKey("j")) {
+ jMap = (Map) signal2DMap.get("j");
+ } else {
+ jMap = null;
+ }
+ if (jMap
+ != null
+ && jMap.containsKey("pathLength")) {
+ pathLengthMap = (Map) jMap.get("pathLength");
+ } else {
+ pathLengthMap = null;
+ }
+
+ for (final int matchIndex : link.getMatch()) {
+ // ignore linked H atoms without any attachment to a heavy atom
+ if (correlationList.get(matchIndex)
+ .getAtomType()
+ .equals("H")
+ && correlationList.get(matchIndex)
+ .getAttachment()
+ .keySet()
+ .isEmpty()) {
+ continue;
+ }
+ if (link.getExperimentType()
+ .equals("hmbc")) {
+ if (molecularConnectivity.getHmbc()
+ == null) {
+ molecularConnectivity.setHmbc(new HashMap<>());
+ }
+ molecularConnectivity.getHmbc()
+ .put(matchIndex, pathLengthMap
+ == null
+ ? defaultBondDistances.get("hmbc")
+ : new Integer[]{(int) pathLengthMap.get("from"),
+ (int) pathLengthMap.get("to")});
+ } else {
+ if (molecularConnectivity.getCosy()
+ == null) {
+ molecularConnectivity.setCosy(new HashMap<>());
+ }
+ molecularConnectivity.getCosy()
+ .put(matchIndex, pathLengthMap
+ == null
+ ? defaultBondDistances.get("cosy")
+ : new Integer[]{(int) pathLengthMap.get("from"),
+ (int) pathLengthMap.get("to")});
+ }
+ }
+ }
+ }
+
+ // set detections
+ if (detections.getForbiddenNeighbors()
+ .containsKey(correlationIndex)) {
+ molecularConnectivity.setForbiddenNeighbors(detections.getForbiddenNeighbors()
+ .get(correlationIndex));
+ }
+ if (detections.getSetNeighbors()
+ .containsKey(correlationIndex)) {
+ molecularConnectivity.setSetNeighbors(detections.getSetNeighbors()
+ .get(correlationIndex));
+ }
+ // fill in fixed neighbors
+ if (correlation.getEquivalence()
+ == 1
+ && detections.getFixedNeighbors()
+ .containsKey(correlationIndex)) {
+ molecularConnectivity.setFixedNeighbors(new ArrayList<>());
+ molecularConnectivity.getFixedNeighbors()
+ .add(correlationIndex);
+ }
+
+ molecularConnectivityList.add(molecularConnectivity);
+ }
+
+ return molecularConnectivityList;
+ }
+
+ public static MolecularConnectivity getHeavyAtomMolecularConnectivity(
+ final Map> molecularConnectivityMap, final int protonIndex) {
+ for (final Map.Entry> entry : molecularConnectivityMap.entrySet()) {
+ for (final MolecularConnectivity molecularConnectivityTemp : entry.getValue()
+ .stream()
+ .filter(mc -> !mc.getAtomType()
+ .equals("H"))
+ .collect(Collectors.toSet())) {
+ if (molecularConnectivityTemp.getHsqc()
+ != null
+ && molecularConnectivityTemp.getHsqc()
+ .contains(protonIndex)) {
+ return molecularConnectivityTemp;
+ }
+ }
+ }
+
+ return null;
+ }
+
+ public static int findBondedHeavyAtomMolecularConnectivityIndex(
+ final List molecularConnectivityList, final int protonIndex) {
+ for (final MolecularConnectivity molecularConnectivityTemp : molecularConnectivityList.stream()
+ .filter(mc -> !mc.getAtomType()
+ .equals("H"))
+ .collect(
+ Collectors.toSet())) {
+ if (molecularConnectivityTemp.getHsqc()
+ != null
+ && molecularConnectivityTemp.getHsqc()
+ .contains(protonIndex)) {
+ return molecularConnectivityTemp.getIndex();
+ }
+ }
+
+ return -1;
+ }
+
+ private static boolean checkDistance(final List molecularConnectivityList, final int index1,
+ final int index2, final Grouping grouping) {
+ final Double distanceValue = casekit.nmr.similarity.Utilities.getDistanceValue(
+ molecularConnectivityList.get(index1)
+ .getSignal(), molecularConnectivityList.get(index2)
+ .getSignal(), 0, 0, false, false, false,
+ grouping.getTolerances()
+ .get("H"));
+
+ return distanceValue
+ != null;
+ }
+
+ private static List swap(final MolecularConnectivity molecularConnectivity, final Grouping grouping,
+ final String experiment,
+ final List molecularConnectivityList,
+ final int correlationIndex, final Set swapped) {
+ final List newStatesList = new ArrayList<>();
+
+ List clonedMolecularConnectivityList;
+ int protonGroupIndex, molecularConnectivityIndexBondedToGroupMember;
+ Set swappedCopy;
+ MolecularConnectivity clonedMolecularConnectivity, clonedMolecularConnectivityBondedToGroupMember;
+ List protonGroupMemberList, indicesToSwap;
+ String swapKey;
+ Integer[] pathLengthProton;
+ if (experiment.equals("hsqc")
+ && molecularConnectivity.getHsqc()
+ != null) {
+ // loop over all matching protons in HSQC
+ for (final Integer protonIndex : molecularConnectivity.getHsqc()) {
+ // only change if it has a group entry
+ if (grouping.getTransformedGroups()
+ .containsKey("H")
+ && grouping.getTransformedGroups()
+ .get("H")
+ .containsKey(protonIndex)) {
+ protonGroupIndex = grouping.getTransformedGroups()
+ .get("H")
+ .get(protonIndex);
+ protonGroupMemberList = grouping.getGroups()
+ .get("H")
+ .get(protonGroupIndex)
+ .stream()
+ .filter(protonGroupMemberIndex -> !Objects.equals(
+ protonGroupMemberIndex, protonIndex)
+ && !molecularConnectivity.getHsqc()
+ .contains(protonGroupMemberIndex))
+ .collect(Collectors.toList());
+ for (final Integer protonGroupMemberIndex : protonGroupMemberList) {
+ indicesToSwap = new ArrayList<>();
+ indicesToSwap.add(protonIndex);
+ indicesToSwap.add(protonGroupMemberIndex);
+ indicesToSwap.sort(Integer::compare); // to always keep the same order
+ swapKey = "hsqc_"
+ + indicesToSwap.stream()
+ .map(String::valueOf)
+ .collect(Collectors.joining("_"));
+ if (swapped.contains(swapKey)
+ || !checkDistance(molecularConnectivityList, protonIndex, protonGroupMemberIndex,
+ grouping)) {
+ continue;
+ }
+ // clone current list element
+ clonedMolecularConnectivity = Utils.cloneObject(molecularConnectivity,
+ MolecularConnectivity.class);
+ // remove the current proton and add the group member proton
+ clonedMolecularConnectivity.getHsqc()
+ .remove(protonIndex);
+ clonedMolecularConnectivity.getHsqc()
+ .add(protonGroupMemberIndex);
+ // copy current list
+ clonedMolecularConnectivityList = new ArrayList<>(molecularConnectivityList);
+ // set cloned and changed list element
+ clonedMolecularConnectivityList.set(correlationIndex, clonedMolecularConnectivity);
+ // check whether group member proton is attached to a heavy atom
+ molecularConnectivityIndexBondedToGroupMember = findBondedHeavyAtomMolecularConnectivityIndex(
+ molecularConnectivityList, protonGroupMemberIndex);
+ if (molecularConnectivityIndexBondedToGroupMember
+ >= 0) {
+ // remove the group member proton from heavy atom and add the current proton
+ clonedMolecularConnectivityBondedToGroupMember = Utils.cloneObject(
+ molecularConnectivityList.get(molecularConnectivityIndexBondedToGroupMember),
+ MolecularConnectivity.class);
+ clonedMolecularConnectivityBondedToGroupMember.getHsqc()
+ .remove(protonGroupMemberIndex);
+ clonedMolecularConnectivityBondedToGroupMember.getHsqc()
+ .add(protonIndex);
+ clonedMolecularConnectivityList.set(molecularConnectivityIndexBondedToGroupMember,
+ clonedMolecularConnectivityBondedToGroupMember);
+ }
+
+ // System.out.println(" --> swapped HSQC at i: "
+ // + molecularConnectivity.getIndex()
+ // + " -> "
+ // + swapKey
+ // + " -> "
+ // + swapped);
+ swappedCopy = new HashSet<>(swapped);
+ swappedCopy.add(swapKey);
+
+ newStatesList.add(new Object[]{clonedMolecularConnectivityList, correlationIndex
+ + 1, swappedCopy});
+ }
+ }
+ }
+ } else if (experiment.equals("hmbc")
+ && molecularConnectivity.getHmbc()
+ != null) {
+
+ // loop over all matching protons in HMBC
+ for (final Integer protonIndex : molecularConnectivity.getHmbc()
+ .keySet()) {
+ // only change if it has a group entry
+ if (grouping.getTransformedGroups()
+ .containsKey("H")
+ && grouping.getTransformedGroups()
+ .get("H")
+ .containsKey(protonIndex)) {
+ protonGroupIndex = grouping.getTransformedGroups()
+ .get("H")
+ .get(protonIndex);
+ protonGroupMemberList = grouping.getGroups()
+ .get("H")
+ .get(protonGroupIndex)
+ .stream()
+ .filter(protonGroupMemberIndex -> !Objects.equals(
+ protonGroupMemberIndex, protonIndex)
+ && (molecularConnectivity.getHsqc()
+ == null
+ || !molecularConnectivity.getHsqc()
+ .contains(protonGroupMemberIndex))
+ && !molecularConnectivity.getHmbc()
+ .containsKey(
+ protonGroupMemberIndex))
+ .collect(Collectors.toList());
+ for (final Integer protonGroupMemberIndex : protonGroupMemberList) {
+ if (!checkDistance(molecularConnectivityList, protonIndex, protonGroupMemberIndex, grouping)) {
+ continue;
+ }
+
+ // copy current list
+ clonedMolecularConnectivityList = new ArrayList<>(molecularConnectivityList);
+ // clone current list element
+ clonedMolecularConnectivity = Utils.cloneObject(molecularConnectivity,
+ MolecularConnectivity.class);
+ // remove the current proton and add the group member proton
+ pathLengthProton = clonedMolecularConnectivity.getHmbc()
+ .get(protonIndex);
+ clonedMolecularConnectivity.getHmbc()
+ .remove(protonIndex);
+ clonedMolecularConnectivity.getHmbc()
+ .putIfAbsent(protonGroupMemberIndex, pathLengthProton);
+ // set cloned and changed list element
+ clonedMolecularConnectivityList.set(correlationIndex, clonedMolecularConnectivity);
+
+ // System.out.println(" --> swapped HMBC at i: "
+ // + molecularConnectivity.getIndex()
+ // + " -> "
+ // + swapped);
+ swappedCopy = new HashSet<>(swapped);
+
+ newStatesList.add(new Object[]{clonedMolecularConnectivityList, correlationIndex
+ + 1, swappedCopy});
+ }
+ }
+ }
+ }
+
+ return newStatesList;
+ }
+
+ private static List> buildCombinations(
+ final List initialMolecularConnectivityList, final Grouping grouping) {
+
+ final List> molecularConnectivityListList = new ArrayList<>();
+ molecularConnectivityListList.add(initialMolecularConnectivityList);
+
+ // final Stack stack = new Stack<>();
+ // stack.push(new Object[]{initialMolecularConnectivityList, 0, new HashSet<>()});
+ //
+ // Object[] objects;
+ // List molecularConnectivityList;
+ // int correlationIndex;
+ // Set swapped;
+ // MolecularConnectivity molecularConnectivity;
+ // List newStateList, newStateList2;
+ // while (!stack.isEmpty()) {
+ // objects = stack.pop();
+ // molecularConnectivityList = (List) objects[0];
+ // correlationIndex = (int) objects[1];
+ // swapped = (Set) objects[2];
+ // if (correlationIndex
+ // >= molecularConnectivityList.size()) {
+ // molecularConnectivityListList.add(molecularConnectivityList);
+ // continue;
+ // }
+ // molecularConnectivity = molecularConnectivityList.get(correlationIndex);
+ // if (!grouping.getGroups()
+ // .containsKey(molecularConnectivity.getAtomType())
+ // || molecularConnectivity.getAtomType()
+ // .equals("H")) {
+ // stack.push(new Object[]{molecularConnectivityList, correlationIndex
+ // + 1, swapped});
+ // continue;
+ // }
+ //
+ // // HSQC
+ // newStateList = swap(molecularConnectivity, grouping, "hsqc", molecularConnectivityList, correlationIndex,
+ // swapped);
+ // // HMBC
+ // newStateList2 = new ArrayList<>(newStateList);
+ // // push unchanged stack data that HMBC can swap from previous data too
+ // newStateList2.add(new Object[]{molecularConnectivityList, correlationIndex, swapped});
+ // for (final Object[] stateObjects : newStateList2) {
+ // newStateList.addAll(
+ // swap(molecularConnectivity, grouping, "hmbc", (List) stateObjects[0],
+ // correlationIndex, (Set) stateObjects[2]));
+ // }
+ //
+ // // push unchanged stack data as well
+ // newStateList.add(new Object[]{molecularConnectivityList, correlationIndex
+ // + 1, swapped});
+ //
+ // // push all new states into the stack
+ // for (final Object[] newStateObjects : newStateList) {
+ // stack.push(newStateObjects);
+ // }
+ // }
+
+
+ return molecularConnectivityListList;
+ }
+
+ public static List>> buildMolecularConnectivityMapCombinationList(
+ final List correlationList, final Detections detections, final Grouping grouping,
+ final Map