List of usage examples for org.apache.commons.lang3.tuple Pair getRight
public abstract R getRight();
Gets the right element from this pair.
When treated as a key-value pair, this is the value.
From source file:act.installer.pubchem.PubchemSynonymFinder.java
public static void main(String[] args) throws Exception { org.apache.commons.cli.Options opts = new org.apache.commons.cli.Options(); for (Option.Builder b : OPTION_BUILDERS) { opts.addOption(b.build());/*from w ww . ja va 2 s . co m*/ } CommandLine cl = null; try { CommandLineParser parser = new DefaultParser(); cl = parser.parse(opts, args); } catch (ParseException e) { System.err.format("Argument parsing failed: %s\n", e.getMessage()); HELP_FORMATTER.printHelp(PubchemSynonymFinder.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } if (cl.hasOption("help")) { HELP_FORMATTER.printHelp(PubchemSynonymFinder.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); return; } File rocksDBFile = new File(cl.getOptionValue(OPTION_INDEX_PATH)); if (!rocksDBFile.isDirectory()) { System.err.format("Index directory does not exist or is not a directory at '%s'", rocksDBFile.getAbsolutePath()); HELP_FORMATTER.printHelp(PubchemSynonymFinder.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } List<String> compoundIds = null; if (cl.hasOption(OPTION_PUBCHEM_COMPOUND_ID)) { compoundIds = Collections.singletonList(cl.getOptionValue(OPTION_PUBCHEM_COMPOUND_ID)); } else if (cl.hasOption(OPTION_IDS_FILE)) { File idsFile = new File(cl.getOptionValue(OPTION_IDS_FILE)); if (!idsFile.exists()) { System.err.format("Cannot find Pubchem CIDs file at %s", idsFile.getAbsolutePath()); HELP_FORMATTER.printHelp(PubchemSynonymFinder.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } compoundIds = getCIDsFromFile(idsFile); if (compoundIds.size() == 0) { System.err.format("Found zero Pubchem CIDs to process in file at '%s', exiting", idsFile.getAbsolutePath()); HELP_FORMATTER.printHelp(PubchemSynonymFinder.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } } else { System.err.format("Must specify one of '%s' or '%s'; index is too big to print all synonyms.", OPTION_PUBCHEM_COMPOUND_ID, OPTION_IDS_FILE); HELP_FORMATTER.printHelp(PubchemSynonymFinder.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } // Run a quick check to warn users of malformed ids. compoundIds.forEach(x -> { if (!PC_CID_PATTERN.matcher(x).matches()) { // Use matches() for complete matching. LOGGER.warn("Specified compound id does not match expected format: %s", x); } }); LOGGER.info("Opening DB and searching for %d Pubchem CIDs", compoundIds.size()); Pair<RocksDB, Map<PubchemTTLMerger.COLUMN_FAMILIES, ColumnFamilyHandle>> dbAndHandles = null; Map<String, PubchemSynonyms> results = new LinkedHashMap<>(compoundIds.size()); try { dbAndHandles = PubchemTTLMerger.openExistingRocksDB(rocksDBFile); RocksDB db = dbAndHandles.getLeft(); ColumnFamilyHandle cidToSynonymsCfh = dbAndHandles.getRight() .get(PubchemTTLMerger.COLUMN_FAMILIES.CID_TO_SYNONYMS); for (String cid : compoundIds) { PubchemSynonyms synonyms = null; byte[] val = db.get(cidToSynonymsCfh, cid.getBytes(UTF8)); if (val != null) { ObjectInputStream oi = new ObjectInputStream(new ByteArrayInputStream(val)); // We're relying on our use of a one-value-type per index model here so we can skip the instanceof check. synonyms = (PubchemSynonyms) oi.readObject(); } else { LOGGER.warn("No synonyms available for compound id '%s'", cid); } results.put(cid, synonyms); } } finally { if (dbAndHandles != null) { dbAndHandles.getLeft().close(); } } try (OutputStream outputStream = cl.hasOption(OPTION_OUTPUT) ? new FileOutputStream(cl.getOptionValue(OPTION_OUTPUT)) : System.out) { OBJECT_MAPPER.writerWithDefaultPrettyPrinter().writeValue(outputStream, results); new OutputStreamWriter(outputStream).append('\n'); } LOGGER.info("Done searching for Pubchem synonyms"); }
From source file:com.act.lcms.db.analysis.IonDetectionAnalysis.java
public static void main(String[] args) throws Exception { Options opts = new Options(); for (Option.Builder b : OPTION_BUILDERS) { opts.addOption(b.build());/*from w w w . j a v a 2 s. c om*/ } CommandLine cl = null; try { CommandLineParser parser = new DefaultParser(); cl = parser.parse(opts, args); } catch (ParseException e) { System.err.format("Argument parsing failed: %s", e.getMessage()); HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } if (cl.hasOption("help")) { HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); return; } File lcmsDir = new File(cl.getOptionValue(OPTION_LCMS_FILE_DIRECTORY)); if (!lcmsDir.isDirectory()) { System.err.format("File at %s is not a directory", lcmsDir.getAbsolutePath()); HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } String plottingDirectory = cl.getOptionValue(OPTION_PLOTTING_DIR); // Get include and excluse ions from command line Set<String> includeIons; if (cl.hasOption(OPTION_INCLUDE_IONS)) { includeIons = new HashSet<>(Arrays.asList(cl.getOptionValues(OPTION_INCLUDE_IONS))); LOGGER.info("Including ions in search: %s", StringUtils.join(includeIons, ", ")); } else { includeIons = new HashSet<>(); includeIons.add(DEFAULT_ION); } try (DB db = DB.openDBFromCLI(cl)) { ScanFile.insertOrUpdateScanFilesInDirectory(db, lcmsDir); File inputPredictionCorpus = new File(cl.getOptionValue(OPTION_INPUT_PREDICTION_CORPUS)); Map<Double, Set<Pair<String, String>>> massChargeToChemicalAndIon = constructMassChargeToChemicalIonsFromInputFile( inputPredictionCorpus, includeIons, cl.hasOption(OPTION_LIST_OF_INCHIS_INPUT_FILE)); Pair<Set<Pair<String, Double>>, Map<String, Double>> values = constructFakeNameToMassChargeAndSetOfMassChargePairs( massChargeToChemicalAndIon.keySet()); Set<Pair<String, Double>> searchMZs = values.getLeft(); Map<String, Double> chemIDToMassCharge = values.getRight(); LOGGER.info("The number of mass charges are: %d", searchMZs.size()); Map<ScanData.KIND, List<LCMSWell>> wellTypeToLCMSWells = readInputExperimentalSetup(db, new File(cl.getOptionValue(OPTION_INPUT_POSITIVE_AND_NEGATIVE_CONTROL_WELLS_FILE))); // Get experimental setup ie. positive and negative wells from config file List<LCMSWell> positiveWells = wellTypeToLCMSWells.get(ScanData.KIND.POS_SAMPLE); List<LCMSWell> negativeWells = wellTypeToLCMSWells.get(ScanData.KIND.NEG_CONTROL); LOGGER.info("Number of positive wells is: %d", positiveWells.size()); LOGGER.info("Number of negative wells is: %d", negativeWells.size()); HashMap<Integer, Plate> plateCache = new HashMap<>(); String outputPrefix = cl.getOptionValue(OPTION_OUTPUT_PREFIX); IonDetectionAnalysis<LCMSWell> ionDetectionAnalysis = new IonDetectionAnalysis<LCMSWell>(lcmsDir, positiveWells, negativeWells, plottingDirectory, plateCache, searchMZs, db); ionDetectionAnalysis.runLCMSMiningAnalysisAndPlotResults(chemIDToMassCharge, massChargeToChemicalAndIon, outputPrefix, cl.hasOption(OPTION_NON_REPLICATE_ANALYSIS)); } }
From source file:com.act.lcms.db.analysis.StandardIonAnalysis.java
public static void main(String[] args) throws Exception { Options opts = new Options(); for (Option.Builder b : OPTION_BUILDERS) { opts.addOption(b.build());/* w w w. ja va 2 s .co m*/ } CommandLine cl = null; try { CommandLineParser parser = new DefaultParser(); cl = parser.parse(opts, args); } catch (ParseException e) { System.err.format("Argument parsing failed: %s\n", e.getMessage()); HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } if (cl.hasOption("help")) { HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); return; } File lcmsDir = new File(cl.getOptionValue(OPTION_DIRECTORY)); if (!lcmsDir.isDirectory()) { System.err.format("File at %s is not a directory\n", lcmsDir.getAbsolutePath()); HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } try (DB db = DB.openDBFromCLI(cl)) { ScanFile.insertOrUpdateScanFilesInDirectory(db, lcmsDir); StandardIonAnalysis analysis = new StandardIonAnalysis(); HashMap<Integer, Plate> plateCache = new HashMap<>(); String plateBarcode = cl.getOptionValue(OPTION_STANDARD_PLATE_BARCODE); String inputChemicals = cl.getOptionValue(OPTION_STANDARD_CHEMICAL); String medium = cl.getOptionValue(OPTION_MEDIUM); // If standard chemical is specified, do standard LCMS ion selection analysis if (inputChemicals != null && !inputChemicals.equals("")) { String[] chemicals; if (!inputChemicals.contains(",")) { chemicals = new String[1]; chemicals[0] = inputChemicals; } else { chemicals = inputChemicals.split(","); } String outAnalysis = cl.getOptionValue(OPTION_OUTPUT_PREFIX) + "." + CSV_FORMAT; String plottingDirectory = cl.getOptionValue(OPTION_PLOTTING_DIR); String[] headerStrings = { "Molecule", "Plate Bar Code", "LCMS Detection Results" }; CSVPrinter printer = new CSVPrinter(new FileWriter(outAnalysis), CSVFormat.DEFAULT.withHeader(headerStrings)); for (String inputChemical : chemicals) { List<StandardWell> standardWells; Plate queryPlate = Plate.getPlateByBarcode(db, cl.getOptionValue(OPTION_STANDARD_PLATE_BARCODE)); if (plateBarcode != null && medium != null) { standardWells = analysis.getStandardWellsForChemicalInSpecificPlateAndMedium(db, inputChemical, queryPlate.getId(), medium); } else if (plateBarcode != null) { standardWells = analysis.getStandardWellsForChemicalInSpecificPlate(db, inputChemical, queryPlate.getId()); } else { standardWells = analysis.getStandardWellsForChemical(db, inputChemical); } if (standardWells.size() == 0) { throw new RuntimeException("Found no LCMS wells for " + inputChemical); } // Sort in descending order of media where MeOH and Water related media are promoted to the top and // anything derived from yeast media are demoted. Collections.sort(standardWells, new Comparator<StandardWell>() { @Override public int compare(StandardWell o1, StandardWell o2) { if (StandardWell.doesMediaContainYeastExtract(o1.getMedia()) && !StandardWell.doesMediaContainYeastExtract(o2.getMedia())) { return 1; } else { return 0; } } }); Map<StandardWell, StandardIonResult> wellToIonRanking = StandardIonAnalysis .getBestMetlinIonsForChemical(inputChemical, lcmsDir, db, standardWells, plottingDirectory); if (wellToIonRanking.size() != standardWells.size() && !cl.hasOption(OPTION_OVERRIDE_NO_SCAN_FILE_FOUND)) { throw new Exception("Could not find a scan file associated with one of the standard wells"); } for (StandardWell well : wellToIonRanking.keySet()) { LinkedHashMap<String, XZ> snrResults = wellToIonRanking.get(well).getAnalysisResults(); String snrRankingResults = ""; int numResultsToShow = 0; Plate plateForWellToAnalyze = Plate.getPlateById(db, well.getPlateId()); for (Map.Entry<String, XZ> ionToSnrAndTime : snrResults.entrySet()) { if (numResultsToShow > 3) { break; } String ion = ionToSnrAndTime.getKey(); XZ snrAndTime = ionToSnrAndTime.getValue(); snrRankingResults += String.format(ion + " (%.2f SNR at %.2fs); ", snrAndTime.getIntensity(), snrAndTime.getTime()); numResultsToShow++; } String[] resultSet = { inputChemical, plateForWellToAnalyze.getBarcode() + " " + well.getCoordinatesString() + " " + well.getMedia() + " " + well.getConcentration(), snrRankingResults }; printer.printRecord(resultSet); } } try { printer.flush(); printer.close(); } catch (IOException e) { System.err.println("Error while flushing/closing csv writer."); e.printStackTrace(); } } else { // Get the set of chemicals that includes the construct and all it's intermediates Pair<ConstructEntry, List<ChemicalAssociatedWithPathway>> constructAndPathwayChems = analysis .getChemicalsForConstruct(db, cl.getOptionValue(OPTION_CONSTRUCT)); System.out.format("Construct: %s\n", constructAndPathwayChems.getLeft().getCompositionId()); for (ChemicalAssociatedWithPathway pathwayChem : constructAndPathwayChems.getRight()) { System.out.format(" Pathway chem %s\n", pathwayChem.getChemical()); // Get all the standard wells for the pathway chemicals. These wells contain only the // the chemical added with controlled solutions (ie no organism or other chemicals in the // solution) List<StandardWell> standardWells; if (plateBarcode != null) { Plate queryPlate = Plate.getPlateByBarcode(db, cl.getOptionValue(OPTION_STANDARD_PLATE_BARCODE)); standardWells = analysis.getStandardWellsForChemicalInSpecificPlate(db, pathwayChem.getChemical(), queryPlate.getId()); } else { standardWells = analysis.getStandardWellsForChemical(db, pathwayChem.getChemical()); } for (StandardWell wellToAnalyze : standardWells) { List<StandardWell> negativeControls = analysis.getViableNegativeControlsForStandardWell(db, wellToAnalyze); Map<StandardWell, List<ScanFile>> allViableScanFiles = analysis .getViableScanFilesForStandardWells(db, wellToAnalyze, negativeControls); List<String> primaryStandardScanFileNames = new ArrayList<>(); for (ScanFile scanFile : allViableScanFiles.get(wellToAnalyze)) { primaryStandardScanFileNames.add(scanFile.getFilename()); } Plate plate = plateCache.get(wellToAnalyze.getPlateId()); if (plate == null) { plate = Plate.getPlateById(db, wellToAnalyze.getPlateId()); plateCache.put(plate.getId(), plate); } System.out.format(" Standard well: %s @ %s, '%s'%s%s\n", plate.getBarcode(), wellToAnalyze.getCoordinatesString(), wellToAnalyze.getChemical(), wellToAnalyze.getMedia() == null ? "" : String.format(" in %s", wellToAnalyze.getMedia()), wellToAnalyze.getConcentration() == null ? "" : String.format(" @ %s", wellToAnalyze.getConcentration())); System.out.format(" Scan files: %s\n", StringUtils.join(primaryStandardScanFileNames, ", ")); for (StandardWell negCtrlWell : negativeControls) { plate = plateCache.get(negCtrlWell.getPlateId()); if (plate == null) { plate = Plate.getPlateById(db, negCtrlWell.getPlateId()); plateCache.put(plate.getId(), plate); } List<String> negativeControlScanFileNames = new ArrayList<>(); for (ScanFile scanFile : allViableScanFiles.get(negCtrlWell)) { negativeControlScanFileNames.add(scanFile.getFilename()); } System.out.format(" Viable negative: %s @ %s, '%s'%s%s\n", plate.getBarcode(), negCtrlWell.getCoordinatesString(), negCtrlWell.getChemical(), negCtrlWell.getMedia() == null ? "" : String.format(" in %s", negCtrlWell.getMedia()), negCtrlWell.getConcentration() == null ? "" : String.format(" @ %s", negCtrlWell.getConcentration())); System.out.format(" Scan files: %s\n", StringUtils.join(negativeControlScanFileNames, ", ")); // TODO: do something useful with the standard wells and their scan files, and then stop all the printing. } } } } } }
From source file:com.act.lcms.db.analysis.IonSearchAnalysis.java
public static void main(String[] args) throws Exception { Options opts = new Options(); for (Option.Builder b : OPTION_BUILDERS) { opts.addOption(b.build());//w w w. j ava 2s .c o m } CommandLine cl = null; try { CommandLineParser parser = new DefaultParser(); cl = parser.parse(opts, args); } catch (ParseException e) { System.err.format("Argument parsing failed: %s\n", e.getMessage()); HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } if (cl.hasOption("help")) { HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); return; } File lcmsDir = new File(cl.getOptionValue(OPTION_DIRECTORY)); if (!lcmsDir.isDirectory()) { System.err.format("File at %s is not a directory\n", lcmsDir.getAbsolutePath()); HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } Double fontScale = null; if (cl.hasOption("font-scale")) { try { fontScale = Double.parseDouble(cl.getOptionValue("font-scale")); } catch (IllegalArgumentException e) { System.err.format("Argument for font-scale must be a floating point number.\n"); System.exit(1); } } try (DB db = DB.openDBFromCLI(cl)) { Set<String> includeIons = null; if (cl.hasOption("include-ions")) { String[] ionNames = cl.getOptionValues("include-ions"); includeIons = new HashSet<>(Arrays.asList(ionNames)); System.out.format("Including ions in search: %s\n", StringUtils.join(includeIons, ", ")); } Set<String> excludeIons = null; if (cl.hasOption("exclude-ions")) { String[] ionNames = cl.getOptionValues("exclude-ions"); excludeIons = new HashSet<>(Arrays.asList(ionNames)); System.out.format("Excluding ions from search: %s\n", StringUtils.join(excludeIons, ", ")); } Set<Integer> takeSamplesFromPlateIds = null; if (cl.hasOption(OPTION_FILTER_BY_PLATE_BARCODE)) { String[] plateBarcodes = cl.getOptionValues(OPTION_FILTER_BY_PLATE_BARCODE); System.out.format("Considering only sample wells in plates: %s\n", StringUtils.join(plateBarcodes, ", ")); takeSamplesFromPlateIds = new HashSet<>(plateBarcodes.length); for (String plateBarcode : plateBarcodes) { Plate p = Plate.getPlateByBarcode(db, plateBarcode); if (p == null) { System.err.format("WARNING: unable to find plate in DB with barcode %s\n", plateBarcode); } else { takeSamplesFromPlateIds.add(p.getId()); } } // Allow filtering on barcode even if we couldn't find any in the DB. } System.out.format("Loading/updating LCMS scan files into DB\n"); ScanFile.insertOrUpdateScanFilesInDirectory(db, lcmsDir); System.out.format("Processing LCMS scans\n"); Pair<List<LCMSWell>, Set<Integer>> positiveWellsAndPlateIds = Utils.extractWellsAndPlateIds(db, cl.getOptionValues(OPTION_STRAINS), cl.getOptionValues(OPTION_CONSTRUCTS), takeSamplesFromPlateIds, false); List<LCMSWell> positiveWells = positiveWellsAndPlateIds.getLeft(); if (positiveWells.size() == 0) { throw new RuntimeException("Found no LCMS wells for specified strains/constructs"); } // Only take negative samples from the plates where we found the positive samples. Pair<List<LCMSWell>, Set<Integer>> negativeWellsAndPlateIds = Utils.extractWellsAndPlateIds(db, cl.getOptionValues(OPTION_NEGATIVE_STRAINS), cl.getOptionValues(OPTION_NEGATIVE_CONSTRUCTS), positiveWellsAndPlateIds.getRight(), true); List<LCMSWell> negativeWells = negativeWellsAndPlateIds.getLeft(); if (negativeWells == null || negativeWells.size() == 0) { System.err.format("WARNING: no valid negative samples found in same plates as positive samples\n"); } // Extract the reference MZ that will be used in the LCMS trace processing. List<Pair<String, Double>> searchMZs = null; Set<CuratedChemical> standardChemicals = null; List<ChemicalAssociatedWithPathway> pathwayChems = null; if (cl.hasOption(OPTION_SEARCH_MZ)) { // Assume mz can be an FP number of a chemical name. String massStr = cl.getOptionValue(OPTION_SEARCH_MZ); Pair<String, Double> searchMZ = Utils.extractMassFromString(db, massStr); if (searchMZ != null) { searchMZs = Collections.singletonList(searchMZ); } standardChemicals = Utils.extractTargetsForWells(db, positiveWells); } else { CuratedChemical targetChemical = Utils.requireOneTarget(db, positiveWells); if (targetChemical == null) { throw new RuntimeException( "Unable to find a curated chemical entry for specified strains'/constructs' targets. " + "Please specify a chemical name or m/z explicitly or update the curated chemicals list in the DB."); } System.out.format("Using reference M/Z for positive target %s (%f)\n", targetChemical.getName(), targetChemical.getMass()); searchMZs = Collections.singletonList(Pair.of(targetChemical.getName(), targetChemical.getMass())); standardChemicals = Collections.singleton(targetChemical); } // Look up the standard by name, or use the target if none is specified. List<StandardWell> standardWells = null; if (cl.hasOption(OPTION_NO_STANDARD)) { System.err.format("WARNING: skipping standard comparison (no-standard option specified)\n"); standardWells = new ArrayList<>(0); } else if (cl.hasOption(OPTION_STANDARD_WELLS)) { String[] standardCoordinates = cl.getOptionValues(OPTION_STANDARD_WELLS); standardWells = new ArrayList<>(standardCoordinates.length); Plate standardPlate = Plate.getPlateByBarcode(db, cl.getOptionValue(OPTION_STANDARD_PLATE_BARCODE)); List<String> foundCoordinates = new ArrayList<>(standardCoordinates.length); for (String stringCoords : standardCoordinates) { Pair<Integer, Integer> coords = Utils.parsePlateCoordinates(stringCoords); StandardWell well = StandardWell.getInstance().getStandardWellsByPlateIdAndCoordinates(db, standardPlate.getId(), coords.getLeft(), coords.getRight()); if (well == null) { System.err.format("Unable to find standard well at %s [%s]\n", standardPlate.getBarcode(), stringCoords); continue; } standardWells.add(well); foundCoordinates.add(stringCoords); } System.out.format("Using explicitly specified standard wells %s [%s]\n", standardPlate.getBarcode(), StringUtils.join(foundCoordinates, ", ")); } else if (cl.hasOption(OPTION_STANDARD_NAME)) { String standardName = cl.getOptionValue(OPTION_STANDARD_NAME); System.out.format("Using explicitly specified standard %s\n", standardName); standardWells = Collections.singletonList(Utils.extractStandardWellFromPlate(db, cl.getOptionValue(OPTION_STANDARD_PLATE_BARCODE), standardName)); } else if (standardChemicals != null && standardChemicals.size() > 0) { // Default to using the target chemical(s) as a standard if none is specified. standardWells = new ArrayList<>(standardChemicals.size()); for (CuratedChemical c : standardChemicals) { String standardName = c.getName(); System.out.format("Searching for well containing standard %s\n", standardName); standardWells.add(Utils.extractStandardWellFromPlate(db, cl.getOptionValue(OPTION_STANDARD_PLATE_BARCODE), standardName)); } } boolean useFineGrainedMZ = cl.hasOption("fine-grained-mz"); boolean useSNR = cl.hasOption(OPTION_USE_SNR); /* Process the standard, positive, and negative wells, producing ScanData containers that will allow them to be * iterated over for graph writing. */ HashMap<Integer, Plate> plateCache = new HashMap<>(); Pair<List<ScanData<StandardWell>>, Double> allStandardScans = AnalysisHelper.processScans(db, lcmsDir, searchMZs, ScanData.KIND.STANDARD, plateCache, standardWells, useFineGrainedMZ, includeIons, excludeIons, useSNR); Pair<List<ScanData<LCMSWell>>, Double> allPositiveScans = AnalysisHelper.processScans(db, lcmsDir, searchMZs, ScanData.KIND.POS_SAMPLE, plateCache, positiveWells, useFineGrainedMZ, includeIons, excludeIons, useSNR); Pair<List<ScanData<LCMSWell>>, Double> allNegativeScans = AnalysisHelper.processScans(db, lcmsDir, searchMZs, ScanData.KIND.NEG_CONTROL, plateCache, negativeWells, useFineGrainedMZ, includeIons, excludeIons, useSNR); String fmt = "pdf"; String outImg = cl.getOptionValue(OPTION_OUTPUT_PREFIX) + "." + fmt; String outData = cl.getOptionValue(OPTION_OUTPUT_PREFIX) + ".data"; System.err.format("Writing combined scan data to %s and graphs to %s\n", outData, outImg); produceLCMSSearchPlots(lcmsDir, outData, outImg, allStandardScans, allPositiveScans, allNegativeScans, fontScale, useFineGrainedMZ, cl.hasOption(OPTION_USE_HEATMAP), useSNR); } }
From source file:com.act.lcms.db.analysis.PathwayProductAnalysis.java
public static void main(String[] args) throws Exception { Options opts = new Options(); for (Option.Builder b : OPTION_BUILDERS) { opts.addOption(b.build());/*from w w w . ja va 2s . c o m*/ } CommandLine cl = null; try { CommandLineParser parser = new DefaultParser(); cl = parser.parse(opts, args); } catch (ParseException e) { System.err.format("Argument parsing failed: %s\n", e.getMessage()); HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } if (cl.hasOption("help")) { HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); return; } File lcmsDir = new File(cl.getOptionValue(OPTION_DIRECTORY)); if (!lcmsDir.isDirectory()) { System.err.format("File at %s is not a directory\n", lcmsDir.getAbsolutePath()); HELP_FORMATTER.printHelp(LoadPlateCompositionIntoDB.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } Double fontScale = null; if (cl.hasOption("font-scale")) { try { fontScale = Double.parseDouble(cl.getOptionValue("font-scale")); } catch (IllegalArgumentException e) { System.err.format("Argument for font-scale must be a floating point number.\n"); System.exit(1); } } try (DB db = DB.openDBFromCLI(cl)) { Set<Integer> takeSamplesFromPlateIds = null; if (cl.hasOption(OPTION_FILTER_BY_PLATE_BARCODE)) { String[] plateBarcodes = cl.getOptionValues(OPTION_FILTER_BY_PLATE_BARCODE); System.out.format("Considering only sample wells in plates: %s\n", StringUtils.join(plateBarcodes, ", ")); takeSamplesFromPlateIds = new HashSet<>(plateBarcodes.length); for (String plateBarcode : plateBarcodes) { Plate p = Plate.getPlateByBarcode(db, plateBarcode); if (p == null) { System.err.format("WARNING: unable to find plate in DB with barcode %s\n", plateBarcode); } else { takeSamplesFromPlateIds.add(p.getId()); } } // Allow filtering on barcode even if we couldn't find any in the DB. } System.out.format("Loading/updating LCMS scan files into DB\n"); ScanFile.insertOrUpdateScanFilesInDirectory(db, lcmsDir); System.out.format("Processing LCMS scans\n"); Pair<List<LCMSWell>, Set<Integer>> positiveWellsAndPlateIds = Utils.extractWellsAndPlateIds(db, cl.getOptionValues(OPTION_STRAINS), cl.getOptionValues(OPTION_CONSTRUCT), takeSamplesFromPlateIds, false); List<LCMSWell> positiveWells = positiveWellsAndPlateIds.getLeft(); if (positiveWells.size() == 0) { throw new RuntimeException("Found no LCMS wells for specified strains/constructs"); } // Only take negative samples from the plates where we found the positive samples. Pair<List<LCMSWell>, Set<Integer>> negativeWellsAndPlateIds = Utils.extractWellsAndPlateIds(db, cl.getOptionValues(OPTION_NEGATIVE_STRAINS), cl.getOptionValues(OPTION_NEGATIVE_CONSTRUCTS), positiveWellsAndPlateIds.getRight(), true); List<LCMSWell> negativeWells = negativeWellsAndPlateIds.getLeft(); if (negativeWells == null || negativeWells.size() == 0) { System.err.format("WARNING: no valid negative samples found in same plates as positive samples\n"); } // Extract the chemicals in the pathway and their product masses, then look up info on those chemicals List<Pair<ChemicalAssociatedWithPathway, Double>> productMasses = Utils .extractMassesForChemicalsAssociatedWithConstruct(db, cl.getOptionValue(OPTION_CONSTRUCT)); List<Pair<String, Double>> searchMZs = new ArrayList<>(productMasses.size()); List<ChemicalAssociatedWithPathway> pathwayChems = new ArrayList<>(productMasses.size()); for (Pair<ChemicalAssociatedWithPathway, Double> productMass : productMasses) { String chemName = productMass.getLeft().getChemical(); searchMZs.add(Pair.of(chemName, productMass.getRight())); pathwayChems.add(productMass.getLeft()); } System.out.format("Searching for intermediate/side-reaction products:\n"); for (Pair<String, Double> searchMZ : searchMZs) { System.out.format(" %s: %.3f\n", searchMZ.getLeft(), searchMZ.getRight()); } // Look up the standard by name. List<StandardWell> standardWells = new ArrayList<>(); if (cl.hasOption(OPTION_STANDARD_WELLS)) { Plate standardPlate = Plate.getPlateByBarcode(db, cl.getOptionValue(OPTION_STANDARD_PLATE_BARCODE)); Map<Integer, StandardWell> pathwayIdToStandardWell = extractStandardWellsFromOptionsList(db, pathwayChems, cl.getOptionValues(OPTION_STANDARD_WELLS), standardPlate); for (ChemicalAssociatedWithPathway c : pathwayChems) { // TODO: we can avoid this loop. StandardWell well = pathwayIdToStandardWell.get(c.getId()); if (well != null) { standardWells.add(well); } } } else { for (ChemicalAssociatedWithPathway c : pathwayChems) { String standardName = c.getChemical(); System.out.format("Searching for well containing standard %s\n", standardName); List<StandardWell> wells = StandardIonAnalysis.getStandardWellsForChemical(db, c.getChemical()); if (wells != null) { standardWells.addAll(wells); } } } boolean useFineGrainedMZ = cl.hasOption("fine-grained-mz"); boolean useSNR = cl.hasOption(OPTION_USE_SNR); /* Process the standard, positive, and negative wells, producing ScanData containers that will allow them to be * iterated over for graph writing. We do not need to specify granular includeIons and excludeIons since * this would not take advantage of our caching strategy which uses a list of metlin ions as an index. */ HashMap<Integer, Plate> plateCache = new HashMap<>(); Pair<List<ScanData<StandardWell>>, Double> allStandardScans = AnalysisHelper.processScans(db, lcmsDir, searchMZs, ScanData.KIND.STANDARD, plateCache, standardWells, useFineGrainedMZ, EMPTY_SET, EMPTY_SET, useSNR); Pair<List<ScanData<LCMSWell>>, Double> allPositiveScans = AnalysisHelper.processScans(db, lcmsDir, searchMZs, ScanData.KIND.POS_SAMPLE, plateCache, positiveWells, useFineGrainedMZ, EMPTY_SET, EMPTY_SET, useSNR); Pair<List<ScanData<LCMSWell>>, Double> allNegativeScans = AnalysisHelper.processScans(db, lcmsDir, searchMZs, ScanData.KIND.NEG_CONTROL, plateCache, negativeWells, useFineGrainedMZ, EMPTY_SET, EMPTY_SET, useSNR); String fmt = "pdf"; String outImg = cl.getOptionValue(OPTION_OUTPUT_PREFIX) + "." + fmt; String outData = cl.getOptionValue(OPTION_OUTPUT_PREFIX) + ".data"; String outAnalysis = cl.getOptionValue(OPTION_OUTPUT_PREFIX) + ".tsv"; System.err.format("Writing combined scan data to %s and graphs to %s\n", outData, outImg); String plottingDirectory = cl.getOptionValue(OPTION_PLOTTING_DIR); List<ScanData<LCMSWell>> posNegWells = new ArrayList<>(); posNegWells.addAll(allPositiveScans.getLeft()); posNegWells.addAll(allNegativeScans.getLeft()); Map<Integer, String> searchIons; if (cl.hasOption(OPTION_PATHWAY_SEARCH_IONS)) { searchIons = extractPathwayStepIons(pathwayChems, cl.getOptionValues(OPTION_PATHWAY_SEARCH_IONS), cl.getOptionValue(OPTION_SEARCH_ION, "M+H")); /* This is pretty lazy, but works with the existing API. Extract all selected ions for all search masses when * performing the scan, then filter down to the desired ions for the plot at the end. * TODO: specify the masses and scans per sample rather than batching everything together. It might be slower, * but it'll be clearer to read. */ } else { // We need to make sure that the standard metlin ion we choose is consistent with the ion modes of // the given positive, negative and standard scan files. For example, we should not pick a negative // metlin ion if all our available positive control scan files are in the positive ion mode. Map<Integer, Pair<Boolean, Boolean>> ionModes = new HashMap<>(); for (ChemicalAssociatedWithPathway chemical : pathwayChems) { boolean isPositiveScanPresent = false; boolean isNegativeScanPresent = false; for (ScanData<StandardWell> scan : allStandardScans.getLeft()) { if (chemical.getChemical().equals(scan.getWell().getChemical()) && chemical.getChemical().equals(scan.getTargetChemicalName())) { if (MS1.IonMode.valueOf( scan.getScanFile().getMode().toString().toUpperCase()) == MS1.IonMode.POS) { isPositiveScanPresent = true; } if (MS1.IonMode.valueOf( scan.getScanFile().getMode().toString().toUpperCase()) == MS1.IonMode.NEG) { isNegativeScanPresent = true; } } } for (ScanData<LCMSWell> scan : posNegWells) { if (chemical.getChemical().equals(scan.getWell().getChemical()) && chemical.getChemical().equals(scan.getTargetChemicalName())) { if (MS1.IonMode.valueOf( scan.getScanFile().getMode().toString().toUpperCase()) == MS1.IonMode.POS) { isPositiveScanPresent = true; } if (MS1.IonMode.valueOf( scan.getScanFile().getMode().toString().toUpperCase()) == MS1.IonMode.NEG) { isNegativeScanPresent = true; } } } ionModes.put(chemical.getId(), Pair.of(isPositiveScanPresent, isNegativeScanPresent)); } // Sort in descending order of media where MeOH and Water related media are promoted to the top and // anything derived from yeast media are demoted. We do this because we want to first process the water // and meoh media before processing the yeast media since the yeast media depends on the analysis of the former. Collections.sort(standardWells, new Comparator<StandardWell>() { @Override public int compare(StandardWell o1, StandardWell o2) { if (StandardWell.doesMediaContainYeastExtract(o1.getMedia()) && !StandardWell.doesMediaContainYeastExtract(o2.getMedia())) { return 1; } else { return 0; } } }); searchIons = extractPathwayStepIonsFromStandardIonAnalysis(pathwayChems, lcmsDir, db, standardWells, plottingDirectory, ionModes); } produceLCMSPathwayHeatmaps(lcmsDir, outData, outImg, outAnalysis, pathwayChems, allStandardScans, allPositiveScans, allNegativeScans, fontScale, cl.hasOption(OPTION_USE_HEATMAP), searchIons); } }
From source file:com.act.analysis.similarity.SubstructureSearch.java
public static void main(String[] args) throws Exception { Options opts = new Options(); for (Option.Builder b : OPTION_BUILDERS) { opts.addOption(b.build());/*from w ww .j a v a 2 s . c o m*/ } CommandLine cl = null; try { CommandLineParser parser = new DefaultParser(); cl = parser.parse(opts, args); } catch (ParseException e) { System.err.format("Argument parsing failed: %s\n", e.getMessage()); HELP_FORMATTER.printHelp(SubstructureSearch.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } if (cl.hasOption("help")) { HELP_FORMATTER.printHelp(SubstructureSearch.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); return; } if (cl.hasOption(OPTION_LICENSE_FILE)) { LicenseManager.setLicenseFile(cl.getOptionValue(OPTION_LICENSE_FILE)); } List<String> searchOpts = Collections.emptyList(); if (cl.hasOption(OPTION_SEARCH_OPTIONS)) { searchOpts = Arrays.asList(cl.getOptionValues(OPTION_SEARCH_OPTIONS)); } // Make sure we can initialize correctly before opening any file handles for writing. SubstructureSearch matcher = new SubstructureSearch(); try { matcher.init(cl.getOptionValue(OPTION_QUERY), searchOpts); } catch (IllegalArgumentException e) { System.err.format("Unable to initialize substructure search. %s\n", e.getMessage()); HELP_FORMATTER.printHelp(SubstructureSearch.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } catch (MolFormatException e) { System.err.format("Invalid SMILES structure query. %s\n", e.getMessage()); HELP_FORMATTER.printHelp(SubstructureSearch.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } Pair<List<String>, Iterator<Map<String, String>>> iterPair = null; if (cl.hasOption(OPTION_INPUT_FILE)) { File inFile = new File(cl.getOptionValue(OPTION_INPUT_FILE)); if (!inFile.exists()) { System.err.format("File at %s does not exist", inFile.getAbsolutePath()); HELP_FORMATTER.printHelp(SubstructureSearch.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } iterPair = iterateOverTSV(inFile); } else if (cl.hasOption(OPTION_INPUT_DB)) { iterPair = iterateOverDB(cl.getOptionValue(OPTION_INPUT_DB_HOST, DEFAULT_HOST), Integer.parseInt(cl.getOptionValue(OPTION_INPUT_DB_HOST, DEFAULT_PORT)), cl.getOptionValue(OPTION_INPUT_DB)); } else { System.err.format("Must specify either input TSV file or input DB from which to read.\n"); HELP_FORMATTER.printHelp(SubstructureSearch.class.getCanonicalName(), HELP_MESSAGE, opts, null, true); System.exit(1); } TSVWriter<String, String> writer = new TSVWriter<>(iterPair.getLeft()); writer.open(new File(cl.getOptionValue(OPTION_OUTPUT_FILE))); LOGGER.info("Seaching for substructure '%s'", cl.getOptionValue(OPTION_QUERY)); try { int rowNum = 0; while (iterPair.getRight().hasNext()) { Map<String, String> row = iterPair.getRight().next(); rowNum++; try { String inchi = row.get(FIELD_INCHI); Molecule target = null; try { target = MolImporter.importMol(inchi); } catch (Exception e) { LOGGER.warn("Skipping molecule %d due to exception: %s\n", rowNum, e.getMessage()); continue; } if (matcher.matchSubstructure(target)) { writer.append(row); writer.flush(); } else { // Don't output if not a match. LOGGER.debug("Found non-matching molecule: %s", inchi); } } catch (SearchException e) { LOGGER.error("Exception on input line %d: %s\n", rowNum, e.getMessage()); throw e; } } } finally { writer.close(); } LOGGER.info("Done with substructure search"); }
From source file:edu.umd.umiacs.clip.tools.math.Formatter.java
public static Pair<Double, Double> format(Pair<Double, Double> pair) { return Pair.of(format(pair.getLeft()), format(pair.getRight())); }
From source file:edu.umd.umiacs.clip.tools.math.CorrelationUtils.java
public static double tauAP(final double[] xUnsorted, final double[] yUnsorted) { Pair<double[], double[]> pairs = sort(xUnsorted, yUnsorted); double[] x = pairs.getLeft(); double[] y = pairs.getRight(); return (2 * range(0, x.length).parallel() .mapToDouble(i -> range(0, x.length).parallel().filter(j -> x[j] > x[i] && y[j] > y[i]).count() / (double) range(0, x.length).parallel().filter(j -> x[j] > x[i]).count()) .filter(Double::isFinite).sum() / (x.length - 1)) - 1; }
From source file:edu.umd.umiacs.clip.tools.math.CorrelationUtils.java
public static double tauGAP(final double[] xUnsorted, final double[] yUnsorted) { Pair<double[], double[]> pairs = sort(xUnsorted, yUnsorted); double[] x = pairs.getLeft(); double[] y = pairs.getRight(); return (2 * range(0, x.length).parallel() .mapToDouble(i -> range(0, x.length).parallel().filter(j -> x[j] > x[i] && y[j] > y[i]) .mapToDouble(j -> max(0, x[j] - x[i])).sum() / range(0, x.length).parallel().mapToDouble(j -> max(0, x[j] - x[i])).sum()) .filter(Double::isFinite).sum() / (x.length - 1)) - 1; }
From source file:com.github.steveash.jg2p.Grams.java
public static Iterable<Pair<String, String>> wordPairsToAllGrams(Iterable<? extends Pair<Word, Word>> words, final GramOptions opts) { return concat(transform(words, new Function<Pair<Word, Word>, Iterable<Pair<String, String>>>() { @Override/*www . j ava 2 s .c o m*/ public Iterable<Pair<String, String>> apply(Pair<Word, Word> input) { return gramProduct(input.getLeft(), input.getRight(), opts); } })); }