package jalview.io;
+import jalview.api.AlignExportSettingI;
+import jalview.api.AlignmentViewPanel;
import jalview.datamodel.HMMNode;
import jalview.datamodel.HiddenMarkovModel;
import jalview.datamodel.SequenceI;
import java.io.BufferedReader;
-import java.io.FileNotFoundException;
import java.io.IOException;
-import java.io.PrintWriter;
-import java.io.UnsupportedEncodingException;
import java.util.ArrayList;
import java.util.List;
import java.util.Scanner;
/**
- * reads in and writes out a HMMER standard file
+ * Adds capability to read in and write out HMMER3 files. .
*
*
* @author TZVanaalten
public class HMMFile extends AlignFile
implements AlignmentFileReaderI, AlignmentFileWriterI
{
- // HMM to store file data
- private HiddenMarkovModel hmm = new HiddenMarkovModel();
+ private static final String TERMINATOR = "//";
+ /*
+ * keys to data in HMM file, used to store as properties of the HiddenMarkovModel
+ */
+ public static final String HMM = "HMM";
+
+ public static final String NAME = "NAME";
+
+ public static final String ACCESSION_NUMBER = "ACC";
+
+ public static final String DESCRIPTION = "DESC";
+
+ public static final String LENGTH = "LENG";
+
+ public static final String MAX_LENGTH = "MAXL";
+
+ public static final String ALPHABET = "ALPH";
+
+ public static final String DATE = "DATE";
+
+ public static final String COMMAND_LOG = "COM";
+
+ public static final String NUMBER_OF_SEQUENCES = "NSEQ";
+
+ public static final String EFF_NUMBER_OF_SEQUENCES = "EFFN";
+
+ public static final String CHECK_SUM = "CKSUM";
+
+ public static final String STATISTICS = "STATS";
+
+ public static final String COMPO = "COMPO";
+
+ public static final String GATHERING_THRESHOLD = "GA";
+
+ public static final String TRUSTED_CUTOFF = "TC";
+
+ public static final String NOISE_CUTOFF = "NC";
+
+ public static final String VITERBI = "VITERBI";
+
+ public static final String MSV = "MSV";
+
+ public static final String FORWARD = "FORWARD";
+
+ public static final String MAP = "MAP";
+
+ public static final String REFERENCE_ANNOTATION = "RF";
+
+ public static final String CONSENSUS_RESIDUE = "CONS";
+
+ public static final String CONSENSUS_STRUCTURE = "CS";
+
+ public static final String MASKED_VALUE = "MM";
+
+ private static final String ALPH_AMINO = "amino";
+ private static final String ALPH_DNA = "DNA";
+ private static final String ALPH_RNA = "RNA";
- // number of possible transitions
- private final int NUMBER_OF_TRANSITIONS = 7;
+ private static final String ALPHABET_AMINO = "ACDEFGHIKLMNPQRSTVWY";
- private final String NEW_LINE = "\n";
+ private static final String ALPHABET_DNA = "ACGT";
+ private static final String ALPHABET_RNA = "ACGU";
- // file header
- String fileHeader;
+ private static final int NUMBER_OF_TRANSITIONS = 7;
- int numberOfSymbols;
+ private static final String SPACE = " ";
- private final String SPACE = " ";
+ /*
+ * optional guide line added to an output HMMER file, purely for readability
+ */
+ private static final String TRANSITIONTYPELINE = " m->m m->i m->d i->m i->i d->m d->d";
+
+ private static String NL = System.lineSeparator();
- private final String COMPO = "COMPO";
+ private HiddenMarkovModel hmm;
- private final String EMPTY = "";
+ // number of symbols in the alphabet used in the hidden Markov model
+ private int numberOfSymbols;
+ /**
+ * Constructor that parses immediately
+ *
+ * @param inFile
+ * @param type
+ * @throws IOException
+ */
+ public HMMFile(String inFile, DataSourceType type) throws IOException
+ {
+ super(inFile, type);
+ }
+
+ /**
+ * Constructor that parses immediately
+ *
+ * @param source
+ * @throws IOException
+ */
public HMMFile(FileParse source) throws IOException
{
- super(false, source);
+ super(source);
}
+ /**
+ * Default constructor
+ */
public HMMFile()
{
-
}
- public HiddenMarkovModel getHMM()
+ /**
+ * Constructor for HMMFile used for exporting
+ *
+ * @param hmm
+ */
+ public HMMFile(HiddenMarkovModel markov)
{
- return hmm;
+ hmm = markov;
}
- public void setHMM(HiddenMarkovModel model)
+ /**
+ * Returns the HMM produced by parsing a HMMER3 file
+ *
+ * @return
+ */
+ public HiddenMarkovModel getHMM()
{
- this.hmm = model;
+ return hmm;
}
+ /**
+ * Gets the name of the hidden Markov model
+ *
+ * @return
+ */
public String getName()
{
return hmm.getName();
}
/**
- * reads data from HMM file
- *
- * @throws IOException
+ * Reads the data from HMM file into the HMM model
*/
@Override
- public void parse() throws IOException
+ public void parse()
{
- parseFileProperties(dataIn);
- parseModel(dataIn);
+ try
+ {
+ hmm = new HiddenMarkovModel();
+ parseHeaderLines(dataIn);
+ parseModel(dataIn);
+ } catch (Exception e)
+ {
+ e.printStackTrace();
+ }
}
-
-
/**
- * imports file properties from hmm file
+ * Reads the header properties from a HMMER3 file and saves them in the
+ * HiddeMarkovModel. This method exits after reading the next line after the
+ * HMM line.
*
* @param input
- * buffered reader used to read in file
* @throws IOException
*/
- void parseFileProperties(BufferedReader input) throws IOException
+ void parseHeaderLines(BufferedReader input) throws IOException
{
- boolean readingFile = true;
- fileHeader = input.readLine();
+ boolean readingHeaders = true;
+ hmm.setFileHeader(input.readLine());
String line = input.readLine();
- while (readingFile)
+ while (readingHeaders && line != null)
{
- if (line != null)
+ Scanner parser = new Scanner(line);
+ String next = parser.next();
+ if (ALPHABET.equals(next))
{
- Scanner parser = new Scanner(line);
- String next = parser.next();
- if ("HMM".equals(next)) // indicates start of HMM data (end of file
- // properties)
- {
- readingFile = false;
- hmm.fillSymbols(parser);
- numberOfSymbols = hmm.getNumberOfSymbols();
- }
- else if ("STATS".equals(next))
- {
- parser.next();
- String key;
- String value;
- key = parser.next();
- value = parser.next() + SPACE + SPACE + parser.next();
- hmm.addFileProperty(key, value);
- }
- else
+ String alphabetType = parser.next();
+ hmm.setProperty(ALPHABET, alphabetType);
+ String alphabet = ALPH_DNA.equalsIgnoreCase(alphabetType)
+ ? ALPHABET_DNA
+ : (ALPH_RNA.equalsIgnoreCase(alphabetType) ? ALPHABET_RNA
+ : ALPHABET_AMINO);
+ numberOfSymbols = hmm.setAlphabet(alphabet);
+ }
+ else if (HMM.equals(next))
+ {
+ readingHeaders = false;
+ String symbols = line.substring(line.indexOf(HMM) + HMM.length());
+ numberOfSymbols = hmm.setAlphabet(symbols);
+ }
+ else if (STATISTICS.equals(next))
+ {
+ parser.next();
+ String key;
+ String value;
+ key = parser.next();
+ value = parser.next() + SPACE + SPACE + parser.next();
+ hmm.setProperty(key, value);
+ }
+ else
+ {
+ String key = next;
+ String value = parser.next();
+ while (parser.hasNext())
{
- String key = next;
- String value = parser.next();
- while (parser.hasNext())
- {
- value = value + SPACE + parser.next();
- }
- hmm.addFileProperty(key, value);
+ value = value + SPACE + parser.next();
}
- parser.close();
+ hmm.setProperty(key, value);
}
+ parser.close();
line = input.readLine();
- if (line == null)
- {
- readingFile = false;
- }
}
-
}
/**
- * parses the model data from the hmm file
+ * Parses the model data from the HMMER3 file. The input buffer should be
+ * positioned at the (optional) COMPO line if there is one, else at the insert
+ * emissions line for the BEGIN node of the model.
*
* @param input
- * buffered reader used to read file
* @throws IOException
*/
void parseModel(BufferedReader input) throws IOException
{
- for (int i = 0; i < hmm.getLength() + 1; i++)
+ /*
+ * specification says there must always be an HMM header (already read)
+ * and one more header (guide headings) which is skipped here
+ */
+ int nodeNo = 0;
+ String line = input.readLine();
+ List<HMMNode> nodes = new ArrayList<>();
+
+ while (line != null && !TERMINATOR.equals(line))
{
- hmm.getNodes().add(new HMMNode());
- String next;
- String line;
- line = input.readLine();
- Scanner matchReader = new Scanner(line);
- next = matchReader.next();
- if (next.equals(COMPO) || i > 0)
+ HMMNode node = new HMMNode();
+ nodes.add(node);
+ Scanner scanner = new Scanner(line);
+ String next = scanner.next();
+
+ /*
+ * expect COMPO (optional) for average match emissions
+ * or a node number followed by node's match emissions
+ */
+ if (COMPO.equals(next) || nodeNo > 0)
{
- // stores match emission line in list
- List<Double> matches = new ArrayList<>();
- matches = fillList(matchReader, numberOfSymbols);
- hmm.getNodes().get(i).setMatchEmissions(matches);
- if (i > 0)
+ /*
+ * parse match emissions
+ */
+ double[] matches = parseDoubles(scanner, numberOfSymbols);
+ node.setMatchEmissions(matches);
+ if (!COMPO.equals(next))
{
- parseAnnotations(matchReader, i);
+ int resNo = parseAnnotations(scanner, node);
+ if (resNo == 0)
+ {
+ /*
+ * no MAP annotation provided, just number off from 0 (begin node)
+ */
+ resNo = nodeNo;
+ }
+ node.setResidueNumber(resNo);
}
+ line = input.readLine();
}
- matchReader.close();
- // stores insert emission line in list
+ scanner.close();
+
+ /*
+ * parse insert emissions
+ */
+ scanner = new Scanner(line);
+ double[] inserts = parseDoubles(scanner, numberOfSymbols);
+ node.setInsertEmissions(inserts);
+ scanner.close();
+
+ /*
+ * parse state transitions
+ */
line = input.readLine();
- Scanner insertReader = new Scanner(line);
- List<Double> inserts = new ArrayList<>();
- inserts = fillList(insertReader, numberOfSymbols);
- hmm.getNodes().get(i).setInsertEmissions(inserts);
- insertReader.close();
-
- // stores state transition line in list
+ scanner = new Scanner(line);
+ double[] transitions = parseDoubles(scanner,
+ NUMBER_OF_TRANSITIONS);
+ node.setStateTransitions(transitions);
+ scanner.close();
line = input.readLine();
- Scanner transitionReader = new Scanner(line);
- List<Double> transitions = new ArrayList<>();
- transitions = fillList(transitionReader, NUMBER_OF_TRANSITIONS);
- hmm.getNodes().get(i).setStateTransitions(transitions);
- transitionReader.close();
+
+ nodeNo++;
}
+ hmm.setNodes(nodes);
}
/**
- * parses annotations on match emission line
+ * Parses the annotations on the match emission line and add them to the node.
+ * (See p109 of the HMMER User Guide (V3.1b2) for the specification.) Returns
+ * the residue position that the node maps to, if provided, else zero.
*
* @param scanner
- * scanner which is processing match emission line
- * @param index
- * index of node which is beign scanned
+ * @param node
*/
- void parseAnnotations(Scanner scanner, int index)
+ int parseAnnotations(Scanner scanner, HMMNode node)
{
- if (hmm.mapIsActive())
+ int mapTo = 0;
+
+ /*
+ * map from hmm node to sequence position, if provided
+ */
+ if (scanner.hasNext())
{
- int column;
- column = scanner.nextInt();
- hmm.getNodes().get(index).setAlignmentColumn(column);
- hmm.getNodeLookup().put(column, index);
+ String value = scanner.next();
+ if (!"-".equals(value))
+ {
+ try
+ {
+ mapTo = Integer.parseInt(value);
+ node.setResidueNumber(mapTo);
+ } catch (NumberFormatException e)
+ {
+ // ignore
+ }
+ }
}
- else
+
+ /*
+ * hmm consensus residue if provided, else '-'
+ */
+ if (scanner.hasNext())
{
- scanner.next();
+ node.setConsensusResidue(scanner.next().charAt(0));
}
- char consensusR;
- consensusR = charValue(scanner.next());
- hmm.getNodes().get(index).setConsensusResidue(consensusR);
-
- char reference;
- reference = charValue(scanner.next());
- hmm.getNodes().get(index).setReferenceAnnotation(reference);
+ /*
+ * RF reference annotation, if provided, else '-'
+ */
+ if (scanner.hasNext())
+ {
+ node.setReferenceAnnotation(scanner.next().charAt(0));
+ }
+ /*
+ * 'm' for masked position, if provided, else '-'
+ */
+ if (scanner.hasNext())
+ {
+ node.setMaskValue(scanner.next().charAt(0));
+ }
- char value;
- value = charValue(scanner.next());
- hmm.getNodes().get(index).setMaskValue(value);
+ /*
+ * structure consensus symbol, if provided, else '-'
+ */
+ if (scanner.hasNext())
+ {
+ node.setConsensusStructure(scanner.next().charAt(0));
+ }
- char consensusS;
- consensusS = charValue(scanner.next());
- hmm.getNodes().get(index).setConsensusStructure(consensusS);
+ return mapTo;
}
-
/**
+ * Fills an array of doubles parsed from an input line
*
* @param input
- * scanner for line containing data to be transferred to list
* @param numberOfElements
- * number of elements in the list to be filled
- * @return filled list
+ * @return
+ * @throws IOException
*/
- static List<Double> fillList(Scanner input,
- int numberOfElements)
+ static double[] parseDoubles(Scanner input,
+ int numberOfElements) throws IOException
{
- List<Double> list = new ArrayList<>();
+ double[] values = new double[numberOfElements];
for (int i = 0; i < numberOfElements; i++)
{
-
+ if (!input.hasNext())
+ {
+ throw new IOException("Incomplete data");
+ }
String next = input.next();
- if (next.contains("*")) // state transitions to or from delete states
- // occasionally have values of -infinity. These
- // values are represented by an * in the .hmm
- // file, and by a null value in the
- // HiddenMarkovModel class
+ if (next.contains("*"))
{
- list.add(Double.NEGATIVE_INFINITY);
+ values[i] = Double.NEGATIVE_INFINITY;
}
else
{
double prob = Double.valueOf(next);
prob = Math.pow(Math.E, -prob);
- list.add(prob);
+ values[i] = prob;
}
}
- return list;
+ return values;
}
-
/**
- * writes a HiddenMarkovModel to a file
+ * Returns a string to be added to the StringBuilder containing the entire
+ * output String.
*
- * @param exportLocation
- * Filename, URL or Pasted String to write to
- * @throws FileNotFoundException
- * @throws UnsupportedEncodingException
- *
- **/
-
- public void exportFile(String exportLocation) throws IOException
- {
- StringBuilder file = new StringBuilder();
- appendFileProperties(file);
- appendModel(file);
- file.append("//");
-
- PrintWriter output = new PrintWriter(exportLocation);
- output.append(file);
- output.close();
-
- }
-
+ * @param initialColumnSeparation
+ * The initial whitespace separation between the left side of the
+ * file and first character.
+ * @param columnSeparation
+ * The separation between subsequent data entries.
+ * @param data
+ * The list of data to be added to the String.
+ * @return
+ */
String addData(int initialColumnSeparation,
int columnSeparation, List<String> data)
{
- String line = EMPTY;
- int index = 0;
+ String line = "";
+ boolean first = true;
for (String value : data)
{
- if (index == 0)
- {
- line += String.format("%" + initialColumnSeparation + "s", value);
- }
- else
- {
- line += String.format("%" + columnSeparation + "s", value);
- }
- index++;
+ int sep = first ? initialColumnSeparation : columnSeparation;
+ line += String.format("%" + sep + "s", value);
+ first = false;
}
return line;
}
+ /**
+ * Converts list of characters into a list of Strings.
+ *
+ * @param list
+ * @return Returns the list of Strings.
+ */
List<String> charListToStringList(List<Character> list)
{
List<String> strList = new ArrayList<>();
return strList;
}
- List<String> doubleListToStringList(List<Double> list,
- int noOfDecimals)
+ /**
+ * Converts an array of doubles into a list of Strings, rounded to the nearest
+ * 5th decimal place
+ *
+ * @param doubles
+ * @param noOfDecimals
+ * @return
+ */
+ List<String> doublesToStringList(double[] doubles)
{
List<String> strList = new ArrayList<>();
- for (double value : list)
+ for (double value : doubles)
{
String strValue;
if (value > 0)
{
strValue = String.format("%.5f", value);
-
}
else if (value == -0.00000d)
{
{
strValue = "*";
}
-
strList.add(strValue);
}
return strList;
}
- List<String> stringArrayToStringList(String[] array)
+ /**
+ * Appends model data in string format to the string builder
+ *
+ * @param output
+ */
+ void appendModelAsString(StringBuilder output)
{
- List<String> list = new ArrayList<>();
- for (String value : array)
+ output.append(HMM).append(" ");
+ String charSymbols = hmm.getSymbols();
+ for (char c : charSymbols.toCharArray())
{
- list.add(value);
+ output.append(String.format("%9s", c));
}
-
- return list;
- }
-
- void appendModel(StringBuilder file)
- {
- String symbolLine = "HMM";
- List<Character> charSymbols = hmm.getSymbols();
- List<String> strSymbols;
- strSymbols = charListToStringList(charSymbols);
- symbolLine += addData(11, 9, strSymbols);
- file.append(symbolLine + NEW_LINE);
-
- String transitionTypeLine = "";
- List<String> transitionTypes;
- transitionTypes = stringArrayToStringList(hmm.getTransitionTypes());
- transitionTypeLine += addData(16, 9, transitionTypes);
- file.append(transitionTypeLine + NEW_LINE);
+ output.append(NL).append(TRANSITIONTYPELINE);
int length = hmm.getLength();
- for (int node = 0; node <= length; node++)
+ for (int nodeNo = 0; nodeNo <= length; nodeNo++)
{
- String matchLine;
- if (node == 0)
- {
- matchLine = String.format("%7s", "COMPO");
- }
- else
- {
- matchLine = String.format("%7s", node);
- }
+ String matchLine = String.format("%7s",
+ nodeNo == 0 ? COMPO : Integer.toString(nodeNo));
- List<String> strMatches;
- List<Double> doubleMatches;
- doubleMatches = hmm.getNode(node).getMatchEmissions();
- convertListToLogSpace(doubleMatches);
- strMatches = doubleListToStringList(doubleMatches, 5);
+ double[] doubleMatches = convertToLogSpace(
+ hmm.getNode(nodeNo).getMatchEmissions());
+ List<String> strMatches = doublesToStringList(doubleMatches);
matchLine += addData(10, 9, strMatches);
-
- if (node != 0)
+ if (nodeNo != 0)
{
- matchLine += SPACE + hmm.getNodeAlignmentColumn(node);
- matchLine += SPACE + hmm.getConsensusResidue(node);
- matchLine += SPACE + hmm.getReferenceAnnotation(node);
- matchLine += SPACE + hmm.getMaskedValue(node);
- matchLine += SPACE + hmm.getConsensusStructure(node);
-
+ matchLine += SPACE + (hmm.getNodeMapPosition(nodeNo));
+ matchLine += SPACE + hmm.getConsensusResidue(nodeNo);
+ matchLine += SPACE + hmm.getReferenceAnnotation(nodeNo);
+ if (hmm.getFileHeader().contains("HMMER3/f"))
+ {
+ matchLine += SPACE + hmm.getMaskedValue(nodeNo);
+ matchLine += SPACE + hmm.getConsensusStructure(nodeNo);
+ }
}
- file.append(matchLine + NEW_LINE);
+ output.append(NL).append(matchLine);
- String insertLine = EMPTY;
- List<String> strInserts;
- List<Double> doubleInserts;
- doubleInserts = hmm.getNode(node).getInsertEmissions();
- convertListToLogSpace(doubleInserts);
- strInserts = doubleListToStringList(doubleInserts, 5);
+ String insertLine = "";
+
+ double[] doubleInserts = convertToLogSpace(
+ hmm.getNode(nodeNo).getInsertEmissions());
+ List<String> strInserts = doublesToStringList(doubleInserts);
insertLine += addData(17, 9, strInserts);
- file.append(insertLine + NEW_LINE);
+ output.append(NL).append(insertLine);
- String transitionLine = EMPTY;
- List<String> strTransitions;
- List<Double> doubleTransitions;
- doubleTransitions = hmm.getNode(node).getStateTransitions();
- convertListToLogSpace(doubleTransitions);
- strTransitions = doubleListToStringList(doubleTransitions, 5);
+ String transitionLine = "";
+ double[] doubleTransitions = convertToLogSpace(
+ hmm.getNode(nodeNo).getStateTransitions());
+ List<String> strTransitions = doublesToStringList(
+ doubleTransitions);
transitionLine += addData(17, 9, strTransitions);
- file.append(transitionLine + NEW_LINE);
+ output.append(NL).append(transitionLine);
}
}
- void appendFileProperties(StringBuilder file)
+ /**
+ * Appends formatted HMM file properties to the string builder
+ *
+ * @param output
+ */
+ void appendProperties(StringBuilder output)
{
- String line;
+ output.append(hmm.getFileHeader());
+
+ String format = "%n%-5s %1s";
+ appendProperty(output, format, NAME);
+ appendProperty(output, format, ACCESSION_NUMBER);
+ appendProperty(output, format, DESCRIPTION);
+ appendProperty(output, format, LENGTH);
+ appendProperty(output, format, MAX_LENGTH);
+ appendProperty(output, format, ALPHABET);
+ appendBooleanProperty(output, format, REFERENCE_ANNOTATION);
+ appendBooleanProperty(output, format, MASKED_VALUE);
+ appendBooleanProperty(output, format, CONSENSUS_RESIDUE);
+ appendBooleanProperty(output, format, CONSENSUS_STRUCTURE);
+ appendBooleanProperty(output, format, MAP);
+ appendProperty(output, format, DATE);
+ appendProperty(output, format, NUMBER_OF_SEQUENCES);
+ appendProperty(output, format, EFF_NUMBER_OF_SEQUENCES);
+ appendProperty(output, format, CHECK_SUM);
+ appendProperty(output, format, GATHERING_THRESHOLD);
+ appendProperty(output, format, TRUSTED_CUTOFF);
+ appendProperty(output, format, NOISE_CUTOFF);
- file.append(fileHeader + NEW_LINE);
-
- line = String.format("%-5s %1s", "NAME", hmm.getName());
- file.append((line + NEW_LINE));
-
- if (hmm.getAccessionNumber() != null)
+ if (hmm.getMSV() != null)
{
- line = String.format("%-5s %1s", "ACC", hmm.getAccessionNumber());
- file.append((line + NEW_LINE));
- }
+ format = "%n%-19s %18s";
+ output.append(String.format(format, "STATS LOCAL MSV", hmm.getMSV()));
- if (hmm.getDescription() != null)
- {
- line = String.format("%-5s %1s", "DESC", hmm.getDescription());
- file.append((line + NEW_LINE));
- }
- line = String.format("%-5s %1s", "LENG", hmm.getLength());
- file.append((line + NEW_LINE));
+ output.append(String.format(format, "STATS LOCAL VITERBI",
+ hmm.getViterbi()));
- if (hmm.getMaxInstanceLength() != null)
- {
- line = String.format("%-5s %1s", "MAXL", hmm.getMaxInstanceLength());
- file.append((line + NEW_LINE));
- }
- line = String.format("%-5s %1s", "ALPH", hmm.getAlphabetType());
- file.append((line + NEW_LINE));
-
- boolean status;
- String statusStr;
-
- status = hmm.referenceAnnotationIsActive();
- statusStr = HiddenMarkovModel.findStringFromBoolean(status);
- line = String.format("%-5s %1s", "RF",
- statusStr);
- file.append((line + NEW_LINE));
-
- status = hmm.maskValueIsActive();
- statusStr = HiddenMarkovModel.findStringFromBoolean(status);
- line = String.format("%-5s %1s", "MM",
- statusStr);
- file.append((line + NEW_LINE));
-
- status = hmm.consensusResidueIsActive();
- statusStr = HiddenMarkovModel.findStringFromBoolean(status);
- line = String.format("%-5s %1s", "CONS",
- statusStr);
- file.append((line + NEW_LINE));
-
- status = hmm.consensusStructureIsActive();
- statusStr = HiddenMarkovModel.findStringFromBoolean(status);
- line = String.format("%-5s %1s", "CS",
- statusStr);
- file.append((line + NEW_LINE));
-
- status = hmm.mapIsActive();
- statusStr = HiddenMarkovModel.findStringFromBoolean(status);
- line = String.format("%-5s %1s", "MAP",
- statusStr);
- file.append((line + NEW_LINE));
-
-
- if (hmm.getDate() != null)
- {
- line = String.format("%-5s %1s", "DATE", hmm.getDate());
- file.append((line + NEW_LINE));
- }
- if (hmm.getNumberOfSequences() != null)
- {
- line = String.format("%-5s %1s", "NSEQ", hmm.getNumberOfSequences());
- file.append((line + NEW_LINE));
- }
- if (hmm.getEffectiveNumberOfSequences() != null)
- {
- line = String.format("%-5s %1s", "EFFN",
- hmm.getEffectiveNumberOfSequences());
- file.append((line + NEW_LINE));
- }
- if (hmm.getCheckSum() != null)
- {
- line = String.format("%-5s %1s", "CKSUM", hmm.getCheckSum());
- file.append((line + NEW_LINE));
- }
- if (hmm.getGatheringThreshold() != null)
- {
- line = String.format("%-5s %1s", "GA", hmm.getGatheringThreshold());
- file.append((line + NEW_LINE));
+ output.append(String.format(format, "STATS LOCAL FORWARD",
+ hmm.getForward()));
}
+ }
- if (hmm.getTrustedCutoff() != null)
+ /**
+ * Appends 'yes' or 'no' for the given property, according to whether or not
+ * it is set in the HMM
+ *
+ * @param output
+ * @param format
+ * @param propertyName
+ */
+ private void appendBooleanProperty(StringBuilder output, String format,
+ String propertyName)
+ {
+ boolean set = hmm.getBooleanProperty(propertyName);
+ output.append(String.format(format, propertyName,
+ set ? HiddenMarkovModel.YES : HiddenMarkovModel.NO));
+ }
+
+ /**
+ * Appends the value of the given property to the output, if not null
+ *
+ * @param output
+ * @param format
+ * @param propertyName
+ */
+ private void appendProperty(StringBuilder output, String format,
+ String propertyName)
+ {
+ String value = hmm.getProperty(propertyName);
+ if (value != null)
{
- line = String.format("%-5s %1s", "TC", hmm.getTrustedCutoff());
- file.append((line + NEW_LINE));
+ output.append(String.format(format, propertyName, value));
}
- if (hmm.getNoiseCutoff() != null)
+ }
+
+ @Override
+ public String print(SequenceI[] sequences, boolean jvsuffix)
+ {
+ if (sequences[0].getHMM() != null)
{
- line = String.format("%-5s %1s", "NC", hmm.getNoiseCutoff());
- file.append((line + NEW_LINE));
+ hmm = sequences[0].getHMM();
}
- if (hmm.getMSV() != null)
+ return print();
+ }
+
+ /**
+ * Prints the .hmm file to a String.
+ *
+ * @return
+ */
+ public String print()
+ {
+ StringBuilder output = new StringBuilder();
+ appendProperties(output);
+ output.append(NL);
+ appendModelAsString(output);
+ output.append(NL).append(TERMINATOR).append(NL);
+ return output.toString();
+ }
+
+ /**
+ * Converts the probabilities contained in an array into log space
+ *
+ * @param ds
+ */
+ double[] convertToLogSpace(double[] ds)
+ {
+ double[] converted = new double[ds.length];
+ for (int i = 0; i < ds.length; i++)
{
- line = String.format("%-19s %18s", "STATS LOCAL MSV", hmm.getMSV());
- file.append((line + NEW_LINE));
-
- line = String.format("%-19s %18s", "STATS LOCAL VITERBI",
- hmm.getViterbi());
- file.append((line + NEW_LINE));
-
- line = String.format("%-19s %18s", "STATS LOCAL FORWARD",
- hmm.getForward());
- file.append((line + NEW_LINE));
+ double prob = ds[i];
+ double logProb = -1 * Math.log(prob);
+
+ converted[i] = logProb;
}
+ return converted;
}
+ /**
+ * Returns the HMM sequence produced by reading a .hmm file.
+ */
+ @Override
+ public SequenceI[] getSeqsAsArray()
+ {
+ SequenceI hmmSeq = hmm.getConsensusSequence();
+ SequenceI[] seq = new SequenceI[1];
+ seq[0] = hmmSeq;
+ return seq;
+ }
-
- char charValue(String string)
+ @Override
+ public void setNewlineString(String newLine)
{
- char character;
- character = string.charAt(0);
- return character;
+ NL = newLine;
}
@Override
- public String print(SequenceI[] seqs, boolean jvsuffix)
+ public void setExportSettings(AlignExportSettingI exportSettings)
{
- return null;
}
- void convertListToLogSpace(List<Double> list)
+ @Override
+ public void configureForView(AlignmentViewPanel viewpanel)
{
- for (int i = 0; i < list.size(); i++)
- {
- double prob = list.get(i);
- double logProb = -1 * Math.log(prob);
-
- list.set(i, logProb);
- }
+ }
+ @Override
+ public boolean hasWarningMessage()
+ {
+ return false;
+ }
+ @Override
+ public String getWarningMessage()
+ {
+ return "warning message";
}
+
}