public ArrayList<String> symbolListToArrayList(SymbolList sl) {
ArrayList<String> result = new ArrayList<String>();
int numberOfSymbols = sl.length();
for (int i = 1; i <= numberOfSymbols; i++) {
String onestate = sl.symbolAt(i).getName();
result.add(onestate);
}
return result;
} /*symbolListToArrayList*/
public String tokenizeSymbolList(SymbolList sl) throws IllegalAlphabetException {
if (sl.getAlphabet() != getAlphabet()) {
throw new IllegalAlphabetException(
"Alphabet " + sl.getAlphabet().getName() + " does not match " + getAlphabet().getName());
}
StringBuffer sb = new StringBuffer();
for (Iterator i = sl.iterator(); i.hasNext(); ) {
Symbol sym = (Symbol) i.next();
try {
Character c = _tokenizeSymbol(sym);
sb.append(c.charValue());
} catch (IllegalSymbolException ex) {
throw new IllegalAlphabetException(ex, "Couldn't tokenize");
}
}
return sb.substring(0);
}
public SymbolList mutate(SymbolList seq)
throws ChangeVetoException, IllegalAlphabetException, IllegalSymbolException {
int maxIndex = getMutationProbs().length - 1;
OrderNDistribution d = getMutationSpectrum();
Random r = new Random();
for (int i = 1; i < seq.length(); i++) {
int index = Math.min(i - 1, maxIndex);
double mutProb = getMutationProbs()[index];
if (r.nextDouble() < mutProb) {
Edit e = new Edit(i, seq.getAlphabet(), d.getDistribution(seq.symbolAt(i)).sampleSymbol());
seq.edit(e);
}
}
return seq;
}
public static void main(String[] args) throws Exception {
// TODO Auto-generated method stub
// TODO Auto-generated method stub
VITERBI3 app = new VITERBI3();
String serDirName = app.getSerDir();
String aliDirName = app.getAliDir();
String[] stateNames = app.getStates();
double[] strProb = app.getStartProbabilities();
double[][] tranProb = app.getTransitionProbability();
app.checkTransitionMatrixNormality(tranProb);
double scoreThr = app.getScoreThreshold();
int minLenForTrhoughs = app.getMinLengthThreshold();
int minLenForPeakDis = app.getMinPeakLengths();
int lengthThresholdForBlocks = app.getBlockLengthThreshold();
double backGroundAlpha[] = {
0.7, 0.8, 1
}; // {0.15106438458196936,0.20973711740901407,0.22900628496706044} {0.8, 0.9, 1}
double greenStateAlpha[] = {
0.9, 2.3, 1
}; // {0.1713002602078479,11.729438447100865,1.3763120488468623}; {0.9,2.3,1}
double greenFollowingMixedAlpha[] = {
0.8, 2.3, 0.9
}; // {0.16327302239493097,9.219505194760753,1.8283941303615243} {0.8,2.3,0.9}
double greenToRedEdgeAlpah[] = {
1.3, 1.4, 1.2
}; // {0.4799631840073914,0.4944929532552769,0.2857067163030646}; {1.3,1.4,1.2}
double redStateAlpha[] = {
2.5, 1, 0.8
}; // {71.25302335371902,0.201959003605188,0.20299272598532336} {5,1.1,0.9}
double redToGreenEdgeAlpha[] = {
5, 1.1, 0.9
}; // {0.39728078185044136,0.3710506533397517,0.2490053255191496}{1.4,1.3,1.2}
// double redFollowingMixedAlpha [] = {10,0.1,0.1};
// //{4.5200496933818615,0.1772646945273629,0.17979061871698085};
double junctionStateAlpha[] = {0.7, 0.8, 1};
LinkedHashMap<String, double[]> statesAndDirPar = new LinkedHashMap<String, double[]>();
// statesAndDirPar.put("P", x);
statesAndDirPar.put("M", backGroundAlpha);
statesAndDirPar.put("r", redStateAlpha);
statesAndDirPar.put("E", redToGreenEdgeAlpha);
statesAndDirPar.put("G", greenStateAlpha);
statesAndDirPar.put("g", greenFollowingMixedAlpha);
statesAndDirPar.put("e", greenToRedEdgeAlpah);
statesAndDirPar.put("R", redStateAlpha);
statesAndDirPar.put("J", junctionStateAlpha);
File serDir = new File(serDirName);
for (File aSerFile : serDir.listFiles()) {
String oneFileName = aSerFile.getName();
if (oneFileName.startsWith(".")) {
continue;
}
BLOCK block = new BLOCK(aSerFile, aliDirName);
String blockId = block.blockId;
int blockLength = app.getBlockLength(blockId);
if (blockLength < lengthThresholdForBlocks) {
// System.out.println(" found sequence :" + blockId + " with too short length :" +
// blockLength);
continue;
}
Matrix2D m = block.blockMatrix;
Sequence seq = block.blockSeq;
ArrayList<Integer> rowsWithSumZero = block.missingData;
SimpleAlphabet observedSeqAlphabet = block.blockObservedSeqAlphabet;
SimpleSymbolList symbolList = block.blockSimpleSymbolList;
// System.out.println("seq " + blockId + " is processed");
MarkovModel mm =
BLOCK.makeMarkovModel(
observedSeqAlphabet, tranProb, strProb, statesAndDirPar, "dirichletMM");
DP dp = new SingleDP(mm);
SymbolList[] symList = {symbolList};
StatePath viterbiPath = dp.viterbi(symList, ScoreType.PROBABILITY);
SymbolList symbolsInViterbi = viterbiPath.symbolListForLabel(StatePath.STATES);
ArrayList<String> viterbiPathAsAnArrayList = app.symbolListToArrayList(symbolsInViterbi);
System.out.println("veterbi path length for " + blockId + " is " + viterbiPath.length());
ArrayList<String> ViterbiPath = new ArrayList<String>();
for (int i = 1; i <= symbolsInViterbi.length(); i++) {
Symbol oneSym = symbolsInViterbi.symbolAt(i);
ViterbiPath.add(oneSym.getName());
// System.out.print(oneSym.getName());
}
System.out.println();
boolean isFiltering = app.getFilterMeaningLessPicks();
app.filterMeaningLessPicks(ViterbiPath, isFiltering);
ArrayList<Double> viterbiTranslatedToNumbersAndMappedToAlingment =
app.mapViterbiPathToAlignment(ViterbiPath, seq, rowsWithSumZero);
String outputFileName =
"/Users/hk3/Desktop/Main/Composure_Droshophila_Model/ANALYSIS_EISENLAB/2L/ViterbiPaths/VITERBI2_"
+ blockId
+ ".txt";
app.printoutViterbiPath(viterbiTranslatedToNumbersAndMappedToAlingment, outputFileName);
SingleDPMatrix forwardMatrix =
(SingleDPMatrix) dp.forwardMatrix(new SymbolList[] {symbolList}, ScoreType.PROBABILITY);
double score = forwardMatrix.getScore();
// System.err.printf("Forward: %g%n", score);
System.out.println();
SingleDPMatrix backwardMatrix =
(SingleDPMatrix) dp.backwardMatrix(new SymbolList[] {symbolList}, ScoreType.PROBABILITY);
ArrayList<Double> posteriorScores =
app.getPosteriorDecodingScores(forwardMatrix, backwardMatrix);
// posteriorScores = app.fillInShortTroughs(posteriorScores, scoreThr, minLenForTrhoughs);
// posteriorScores = app.filterOutShortPeaks(posteriorScores, scoreThr, minLenForPeakDis);
// posteriorScores = app.getEnhancerRegions(posteriorScores, scoreThr);
ArrayList<Double> PDScoresMapedToalignment =
app.mapPDScoresToAlignment(posteriorScores, seq, rowsWithSumZero);
String posteriorOutputFileName =
"/Users/hk3/Desktop/Main/Composure_Droshophila_Model/ANALYSIS_EISENLAB/2L/ViterbiPaths/Posterior_"
+ blockId
+ ".txt";
app.printoutViterbiPath(PDScoresMapedToalignment, posteriorOutputFileName);
// writte chains into gff files
String chainDir = app.getChainDir();
app.wirteChainsIntoAGffFile(PDScoresMapedToalignment, blockId, chainDir, scoreThr);
}
} /*main*/
private void dnaCommand(HttpServletRequest req, DazzleResponse resp, DazzleDataSource dds)
throws IOException, DataSourceException, ServletException, DazzleException {
DazzleReferenceSource drs = (DazzleReferenceSource) dds;
List segments = DazzleTools.getSegments(dds, req, resp);
if (segments.size() == 0) {
throw new DazzleException(
DASStatus.STATUS_BAD_COMMAND_ARGUMENTS, "No segments specified for dna command");
}
// Fetch and validate the requests.
Map segmentResults = new HashMap();
for (Iterator i = segments.iterator(); i.hasNext(); ) {
Segment seg = (Segment) i.next();
try {
Sequence seq = drs.getSequence(seg.getReference());
if (seq.getAlphabet() != DNATools.getDNA()) {
throw new DazzleException(
DASStatus.STATUS_SERVER_ERROR,
"Sequence " + seg.toString() + " is not in the DNA alphabet");
}
if (seg.isBounded()) {
if (seg.getMin() < 1 || seg.getMax() > seq.length()) {
throw new DazzleException(
DASStatus.STATUS_BAD_COORDS,
"Segment " + seg.toString() + " doesn't fit sequence of length " + seq.length());
}
}
segmentResults.put(seg, seq);
} catch (NoSuchElementException ex) {
throw new DazzleException(DASStatus.STATUS_BAD_REFERENCE, ex);
} catch (DataSourceException ex) {
throw new DazzleException(DASStatus.STATUS_SERVER_ERROR, ex);
}
}
//
// Looks okay -- generate the response document
//
XMLWriter xw = resp.startDasXML("DASDNA", "dasdna.dtd");
try {
xw.openTag("DASDNA");
for (Iterator i = segmentResults.entrySet().iterator(); i.hasNext(); ) {
Map.Entry me = (Map.Entry) i.next();
Segment seg = (Segment) me.getKey();
Sequence seq = (Sequence) me.getValue();
xw.openTag("SEQUENCE");
xw.attribute("id", seg.getReference());
xw.attribute("version", drs.getLandmarkVersion(seg.getReference()));
if (seg.isBounded()) {
xw.attribute("start", "" + seg.getStart());
xw.attribute("stop", "" + seg.getStop());
} else {
xw.attribute("start", "" + 1);
xw.attribute("stop", "" + seq.length());
}
SymbolList syms = seq;
if (seg.isBounded()) {
syms = syms.subList(seg.getMin(), seg.getMax());
}
if (seg.isInverted()) {
syms = DNATools.reverseComplement(syms);
}
xw.openTag("DNA");
xw.attribute("length", "" + syms.length());
for (int pos = 1; pos <= syms.length(); pos += 60) {
int maxPos = Math.min(syms.length(), pos + 59);
xw.println(syms.subStr(pos, maxPos));
}
xw.closeTag("DNA");
xw.closeTag("SEQUENCE");
}
xw.closeTag("DASDNA");
xw.close();
} catch (Exception ex) {
throw new DazzleException(ex, "Error writing DNA document");
}
}
/**
* Calculate the predicted properties of this polypeptide.
*
* @return a <code>PeptideProperties</code> object containing the predicted properties of this
* polypeptide.
*/
public PeptideProperties calculateStats() {
if (this.getResidues() == null) {
logger.warn("No residues for '" + this.getUniqueName() + "'");
return null;
}
String residuesString = new String(this.getResidues());
SymbolList residuesSymbolList = null;
PeptideProperties pp = new PeptideProperties();
try {
SymbolTokenization proteinTokenization = ProteinTools.getTAlphabet().getTokenization("token");
residuesSymbolList = new SimpleSymbolList(proteinTokenization, residuesString);
if (residuesSymbolList.length() == 0) {
logger.error(
String.format(
"Polypeptide feature '%s' has zero-length residues", this.getUniqueName()));
return pp;
}
try {
// if the sequence ends with a termination symbol (*), we need to remove it
if (residuesSymbolList.symbolAt(residuesSymbolList.length()) == ProteinTools.ter()) {
if (residuesSymbolList.length() == 1) {
logger.error(
String.format(
"Polypeptide feature '%s' only has termination symbol", this.getUniqueName()));
return pp;
}
residuesSymbolList = residuesSymbolList.subList(1, residuesSymbolList.length() - 1);
}
} catch (IndexOutOfBoundsException exception) {
throw new RuntimeException(exception);
}
} catch (BioException e) {
logger.error("Can't translate into a protein sequence", e);
return pp;
}
pp.setAminoAcids(residuesSymbolList.length());
try {
double isoElectricPoint = new IsoelectricPointCalc().getPI(residuesSymbolList, false, false);
pp.setIsoelectricPoint(isoElectricPoint);
} catch (Exception e) {
logger.error(
String.format("Error computing protein isoelectric point for '%s'", residuesSymbolList),
e);
}
double mass2 = calculateMass(residuesSymbolList);
if (mass2 != -1) {
// mass = mass2;
pp.setMass(mass2);
}
double charge = calculateCharge(residuesString);
pp.setCharge(charge);
return pp;
}
/**
* Returns the input subsequence matched by the previous match.
*
* <p>For a matcher m with input sequence s, the expressions m.group() and s.substring(m.start(),
* m.end()) are equivalent. Note that some patterns, for example a*, match the empty SymbolList.
* This method will return the empty string when the pattern successfully matches the empty string
* in the input.
*
* @return The (possibly empty) subsequence matched by the previous match, in SymbolList form.
*/
public SymbolList group() {
return sl.subList(start(), end() - 1);
}
/**
* Returns the input subsequence captured by the given group during the previous match operation.
*
* <p>For a matcher m, input sequence s, and group index g, the expressions m.group(g) and
* s.substring(m.start(g), m.end(g)) are equivalent. Capturing groups are indexed from left to
* right, starting at one. Group zero denotes the entire pattern, so the expression m.group(0) is
* equivalent to m.group(). If the match was successful but the group specified failed to match
* any part of the input sequence, then null is returned. Note that some groups, for example (a*),
* match the empty string. This method will return the empty string when such a group successfully
* matches the emtpy string in the input.
*
* @return The (possibly empty) subsequence captured by the group during the previous match, or
* null if the group failed to match part of the input.
*/
public SymbolList group(int group) throws IndexOutOfBoundsException {
int start = matcher.start(group);
int end = matcher.end(group);
if ((start == -1) && (end == -1)) return null;
else return sl.subList(start(group), end(group) - 1);
}
