standardtmhmmcreator.java

马尔科夫模型的c语言实现

import java.io.*;
import java.util.*;

class StandardTMHMMCreator
{
    

    BufferedReader reader;
    ModelMaker modelmaker;
    String hmmrow;
    

    /* first input parameter is a file with rows consisting of: <hmmname> <hmmsize> [priorfile]
     * second input parameter is a name of outputdir */
    public StandardTMHMMCreator(String hmmfile, String outdir)
    {
	try {
	    BufferedReader br = new BufferedReader(new FileReader(hmmfile));
	    hmmrow = br.readLine();
	    StringTokenizer st = new StringTokenizer(hmmrow, "\t ");
	    if(st.countTokens() < 1) {
		printerr("Could not create hmm");
		System.exit(0);
	    }
	    modelmaker = new ModelMaker();
	    
	    /* read name */
	    String hmmName = st.nextToken();
	    
	    /* read priorfiles */
	    LinkedList priorfiles = new LinkedList();
	    LinkedList priorScalers = new LinkedList();
	    for(int i = 0; i < 7; i++) {
		hmmrow = br.readLine();
		st = new StringTokenizer(hmmrow, " \t");
		st.nextToken();
		priorfiles.add(st.nextToken());
		priorScalers.add(st.nextToken());
	    }

	    /* read init distributions */
	    double[][] emissionProbs = new double[7][20];
	    for(int i = 0; i < 7; i++) {
		hmmrow = br.readLine();
		st = new StringTokenizer(hmmrow, "\t ");
		if(st.countTokens() != 21) {
		    printerr("Could not create hmm");
		    System.exit(0);
		}
		st.nextToken();
		for(int j = 0; j < 20; j++) {
		    emissionProbs[i][j] = Double.parseDouble(st.nextToken());
		}
	    }

	    /* read expected looplengths */
	    hmmrow = br.readLine();
	    st = new StringTokenizer(hmmrow, " \t");
	    st.nextToken();
	    int lengthOL = Integer.parseInt(st.nextToken());
	    hmmrow = br.readLine();
	    st = new StringTokenizer(hmmrow, " \t");
	    st.nextToken();
	    int lengthOS = Integer.parseInt(st.nextToken());
	    hmmrow = br.readLine();
	    st = new StringTokenizer(hmmrow, " \t");
	    st.nextToken();
	    int lengthI = Integer.parseInt(st.nextToken());
	    hmmrow = br.readLine();
	    st = new StringTokenizer(hmmrow, " \t");
	    st.nextToken();
	    double Po = Double.parseDouble(st.nextToken());
	    hmmrow = br.readLine();
	    st = new StringTokenizer(hmmrow, " \t");
	    st.nextToken();
	    int f_i = Integer.parseInt(st.nextToken());
	    int q_i = Integer.parseInt(st.nextToken());
	    hmmrow = br.readLine();
	    st = new StringTokenizer(hmmrow, " \t");
	    st.nextToken();
	    int f_o = Integer.parseInt(st.nextToken());
	    int q_o = Integer.parseInt(st.nextToken());
	    
	    createHMM(hmmName, priorfiles, priorScalers, emissionProbs, lengthOL, lengthOS, lengthI, Po, f_i, q_i, f_o, q_o);
	    saveHMM(outdir);
	}
	catch(IOException e) {
	    printerr("Could not read from hmmfile");
	    System.exit(0);
	}
    }


    /****************input/output handling**********************************/
    private void println(String s)
    {
	System.out.println(s);
    }

    private void newln(){
	System.out.println("");
    }
    private void printerr(String s)
    {
	System.out.println("Error: " + s);
    }
    private void print(String s)
    {
	System.out.print(s);
	System.out.flush();
    }

    private String readln()
    {
	try {
	String s = reader.readLine();
	if(s == "" || s == null) {
	    return null;
	}
	s = s.trim();
	return s;
	}
	catch(IOException e) {
	    P.INTERNAL_ERROR("DataReader.readLine: IOException");
	    return null;
	}
    }
    
    
    /*******************************HMM methods****************************/
    private void createHMM(String name, LinkedList priorfiles, LinkedList priorScalers, double[][] emissionProbs, int lengthOL, int lengthOS, int lengthI,
			   double Po, int f_i, int q_i, int f_o, int q_o)
    {
	modelmaker.createHMM(name);
	modelmaker.setAlphabet(HMM.AMINO);
	int res = modelmaker.createModule("s", HMM.STARTNODE, HMM.ZERO, 1, "0");

	res = modelmaker.createModule("in10", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in11", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in12", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in13", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in14", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in15", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in16", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in17", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in18", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in19", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in20", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in21", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in22", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in23", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in24", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in25", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in26", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in27", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in28", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("in29", HMM.SINGLENODE, HMM.EVEN, 1, "i");
	res = modelmaker.createModule("inglob1", HMM.SINGLELOOP, HMM.EVEN, 1, "i");
	
	res = modelmaker.createModule("outglob10", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob11", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob12", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob13", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob14", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob15", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob16", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob17", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob18", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob19", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob20", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob21", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob22", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob23", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob24", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob25", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob26", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob27", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob28", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglob29", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglobLong", HMM.SINGLELOOP, HMM.EVEN, 1, "o");

	res = modelmaker.createModule("out10", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out11", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out12", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out13", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out14", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out15", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out16", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out17", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out18", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out19", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out20", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out21", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out22", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out23", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out24", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out25", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out26", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out27", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out28", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("out29", HMM.SINGLENODE, HMM.EVEN, 1, "o");
	res = modelmaker.createModule("outglobShort", HMM.SINGLELOOP, HMM.EVEN, 1, "o");

	res = modelmaker.createModule("ohelixi1", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixi2", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixi3", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixi4", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixi5", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixi6", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixi7", HMM.SINGLENODE, HMM.EVEN, 1, "M");

	res = modelmaker.createModule("ohelix", HMM.FORWARD_STD, HMM.EVEN, 2, 22, "M");

	res = modelmaker.createModule("ohelixo2", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixo3", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixo4", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixo5", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixo6", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ohelixo7", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	
	res = modelmaker.createModule("ihelixi2", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixi3", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixi4", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixi5", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixi6", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixi7", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	
	res = modelmaker.createModule("ihelix", HMM.FORWARD_STD, HMM.EVEN, 2, 22, "M");

	res = modelmaker.createModule("ihelixo1", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixo2", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixo3", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixo4", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixo5", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixo6", HMM.SINGLENODE, HMM.EVEN, 1, "M");
	res = modelmaker.createModule("ihelixo7", HMM.SINGLENODE, HMM.EVEN, 1, "M");

	res = modelmaker.createModule("e", HMM.ENDNODE, HMM.ZERO, 1, "0");

	
	
	modelmaker.setTransition("s", "in10");
	modelmaker.setTransition("s", "out10");
	modelmaker.setTransition("s", "outglob10");
	modelmaker.initializeTransitionProbabilities("s");
	
	double distribSum = 0.0;
	for(int i = 1; i <= 20; i++) {
	    distribSum += Math.pow(i, (f_i/2 - 1)) * Math.exp((0-i)/q_i);
	}
	double tot = 1.0;

	modelmaker.setTransition("in10", "in11");
	modelmaker.setTransition("in10", "in29");
	modelmaker.setTransition("in10", "ohelixi1");
	modelmaker.setTransition("in10", "e");
	tot = setChiSquareDistrib("in10", 1, 2, f_i, q_i, distribSum, tot);
	modelmaker.initializeTransitionProbabilities("in10");

	modelmaker.setTransition("in11", "in12");
	modelmaker.setTransition("in11", "in28");
	modelmaker.setTransition("in11", "in29");
	modelmaker.setTransition("in11", "e");
	tot = setChiSquareDistrib("in11", 3, 4, f_i, q_i, distribSum, tot);
	modelmaker.initializeTransitionProbabilities("in11");

	modelmaker.setTransition("in12", "in13");
	modelmaker.setTransition("in12", "in27");
	modelmaker.setTransition("in12", "in28");
	modelmaker.setTransition("in12", "e");
	tot = setChiSquareDistrib("in12", 5, 6, f_i, q_i, distribSum, tot);
	modelmaker.initializeTransitionProbabilities("in12");

	modelmaker.setTransition("in13", "in14");
	modelmaker.setTransition("in13", "in26");
	modelmaker.setTransition("in13", "in27");
	modelmaker.setTransition("in13", "e");
	tot = setChiSquareDistrib("in13", 7, 8, f_i, q_i, distribSum, tot);
	modelmaker.initializeTransitionProbabilities("in13");

	modelmaker.setTransition("in14", "in15");
	modelmaker.setTransition("in14", "in25");
	modelmaker.setTransition("in14", "in26");
	modelmaker.setTransition("in14", "e");
	tot = setChiSquareDistrib("in14", 9, 10, f_i, q_i, distribSum, tot);
	modelmaker.initializeTransitionProbabilities("in14");	
	
	modelmaker.setTransition("in15", "in16");
	modelmaker.setTransition("in15", "in24");
	modelmaker.setTransition("in15", "in25");
	modelmaker.setTransition("in15", "e");
	tot = setChiSquareDistrib("in15", 11, 12, f_i, q_i, distribSum, tot);
	modelmaker.initializeTransitionProbabilities("in15");
	
	modelmaker.setTransition("in16", "in17");
	modelmaker.setTransition("in16", "in23");
	modelmaker.setTransition("in16", "in24");
	modelmaker.setTransition("in16", "e");
	tot = setChiSquareDistrib("in16", 13, 14, f_i, q_i, distribSum, tot);
	modelmaker.initializeTransitionProbabilities("in16");

	modelmaker.setTransition("in17", "in18");
	modelmaker.setTransition("in17", "in22");
	modelmaker.setTransition("in17", "in23");
	modelmaker.setTransition("in17", "e");
	tot = setChiSquareDistrib("in17", 15, 16, f_i, q_i, distribSum, tot);
	modelmaker.initializeTransitionProbabilities("in17");

	modelmaker.setTransition("in18", "in19");
	modelmaker.setTransition("in18", "in21");
	modelmaker.setTransition("in18", "in22");