Package edu.msu.cme.rdp.classifier.train

Examples of edu.msu.cme.rdp.classifier.train.LineageSequenceParser

50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
     * @return
     * @throws IOException
     */
    public static Set<String> randomSelectTaxon(File tax_file, File source_file, float fraction, String rank) throws IOException{
        TreeFactory factory = new TreeFactory(new FileReader(tax_file));
        LineageSequenceParser parser = new LineageSequenceParser(source_file);
        HashMap<String, HashSet> genusTrainSeqMap = new HashMap<String, HashSet>(); // keep the seqID for each genus

        while ( parser.hasNext() ){
            LineageSequence pSeq = parser.next();
            HierarchyTree genusNode = factory.addSequence( pSeq);
            HashSet<String> genusSeqSet = genusTrainSeqMap.get(genusNode.getName());
            if ( genusSeqSet == null){
                genusSeqSet = new HashSet<String>();
                genusTrainSeqMap.put(genusNode.getName(), genusSeqSet);
            }
            genusSeqSet.add(pSeq.getSeqName());
        }
        parser.close();


        // random select nodes at the give rank level
        ArrayList<HierarchyTree> nodeList = new ArrayList<HierarchyTree>();
        factory.getRoot().getNodeList(rank, nodeList);      
View Full Code Here
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
    public void testNext() throws FileNotFoundException, IOException {
        System.err.println("testNext()");
        File in = new File(System.class.getResource("/test/classifier/testSeqParser.fasta").getFile());

        LineageSequenceParser parser = new LineageSequenceParser(in);
        boolean next = parser.hasNext();
        assertTrue(next);
        // test the first sequence
        LineageSequence pSeq = parser.next();
        String name = "X53199";
        assertEquals(name.toUpperCase(), pSeq.getSeqName());
        assertEquals("AZOSPIRILLUM_GROUP", (String) pSeq.getAncestors().get(4));
        assertEquals("PROTEOBACTERIA", (String) pSeq.getAncestors().get(2));
        assertEquals("GAEA", (String) pSeq.getAncestors().get(0));

        //test the last sequence
        parser.next();
        pSeq = parser.next();
        name = "AB002485";
        assertEquals(name.toUpperCase(), pSeq.getSeqName());
        assertEquals("M.RRR_SUBGROUP", (String) pSeq.getAncestors().get(5));
        assertEquals("TTTT_GROUP", (String) pSeq.getAncestors().get(4));
        assertEquals("BELTA_SUBDIVISION", (String) pSeq.getAncestors().get(3));
        assertEquals("GAEA", (String) pSeq.getAncestors().get(0));
        String sequence = "AAAAUAtttAGUCCCCCCCCUG";

        assertEquals(sequence, pSeq.getSeqString());
        assertTrue(!parser.hasNext());
    }
View Full Code Here
42
43
44
45
46
47
48
49
50
51
52
53
54
55
      Reader taxReader = new FileReader(System.class.getResource("/test/classifier/testDMtaxon.txt").getFile());
      TreeFactory factory = new TreeFactory(taxReader);
     
      File infileReader = new File(System.class.getResource("/test/classifier/testDMtrain.fasta").getFile()); 
   
      LineageSequenceParser parser = new LineageSequenceParser(infileReader);    
      
      while ( parser.hasNext() ){
        factory.addSequence( (LineageSequence)parser.next());       
      }  
      //after all the training set is being parsed, calculate the prior probability for all the words.
     
      factory.calculateWordPrior();           
     
View Full Code Here
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
        Taxonomy c2Taxon = new Taxonomy(3,"child2", 1, 1, "GENUS");
        HierarchyTree c2 = new HierarchyTree("child2", root, c2Taxon);

        File in = new File(System.class.getResource("/test/classifier/testTrainingSet.fasta").getFile());

        LineageSequenceParser parser = new LineageSequenceParser(in);
        // test the first sequence
        LineageSequence pSeq = parser.next();
        c1.initWordOccurrence(pSeq, wordPriorArr, true);

        pSeq = parser.next();
        c1.initWordOccurrence(pSeq, wordPriorArr, true);

        pSeq = parser.next();
        c2.initWordOccurrence(pSeq, wordPriorArr, true);

        assertEquals(2, c1.getNumOfLeaves());
        assertEquals(1, c2.getNumOfLeaves());
        assertEquals(3, root.getNumOfLeaves());
View Full Code Here
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
        HierarchyTree c1 = new HierarchyTree("child1", root, c1Taxon);
        Taxonomy c2Taxon = new Taxonomy(3, "child2", 1, 1, "GENUS");
        HierarchyTree c2 = new HierarchyTree("child2", root, c2Taxon);
        File in = new File(System.class.getResource("/test/classifier/testTrainingSet.fasta").getFile());

        LineageSequenceParser parser = new LineageSequenceParser(in);
        // test the first sequence
        LineageSequence pSeq = parser.next();
        c1.initWordOccurrence(pSeq, wordPriorArr, true);

        pSeq = parser.next();
        c1.initWordOccurrence(pSeq, wordPriorArr, true);

        pSeq = parser.next();
        c2.initWordOccurrence(pSeq, wordPriorArr, true);

        assertEquals(2, c1.getNumOfLeaves());
        assertEquals(1, c2.getNumOfLeaves());
        assertEquals(3, root.getNumOfLeaves());
View Full Code Here
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
        TreeFactory factory = new TreeFactory(taxReader);

        File infileReader = new File(System.class.getResource("/test/classifier/testNBClassifierSet.fasta").getFile());

        LineageSequenceParser parser = new LineageSequenceParser(infileReader);

        while (parser.hasNext()) {
            factory.addSequence((LineageSequence) parser.next());
        }
        //after all the training set is being parsed, calculate the prior probability for all the words.

        factory.calculateWordPrior();

        HashMap<String, HierarchyTree> nodeMap = new HashMap<String, HierarchyTree>();
        HierarchyTree root = factory.getRoot();
        root.getNodeMap("GENUS", nodeMap);
        ArrayList<HierarchyTree> nodeList = new ArrayList(nodeMap.values());
       
        // test the first sequence
        File queryReader = new File(System.class.getResource("/test/classifier/testNBClassifierSet.fasta").getFile());       
        parser = new LineageSequenceParser(queryReader);
        LineageSequence pSeq = parser.next();
        GoodWordIterator iterator = new GoodWordIterator(pSeq.getSeqString());
        NBClassifier classifier = new NBClassifier(factory, iterator.getWordArr(), nodeList, useSeed, min_bootstrap_words);

        assertEquals(pSeq.getSeqName(), "XG1_child1");       
        ValidationClassificationResult result = classifier.assignClass();
        assertEquals("G1", ((HierarchyTree) result.getBestClass()).getName());
        assertTrue(0.1 > result.getPosteriorProb());


        pSeq = parser.next();
        // test the 3rd getSequence()
        pSeq = parser.next();
        assertEquals(pSeq.getSeqName(), "XG2_child1");
        iterator = new GoodWordIterator(pSeq.getSeqString());
        classifier = new NBClassifier(factory, iterator.getWordArr(), nodeList, useSeed, min_bootstrap_words);
        result = classifier.assignClass();
        assertEquals("G2", ((HierarchyTree) result.getBestClass()).getName());
        assertTrue(0.2 > result.getPosteriorProb());


        pSeq = parser.next();
        // test the 5th getSequence()
        pSeq = parser.next();
        assertEquals(pSeq.getSeqName(), "XPh2G6_child1");
        iterator = new GoodWordIterator(pSeq.getSeqString());
        classifier = new NBClassifier(factory, iterator.getWordArr(), nodeList, useSeed, min_bootstrap_words);
        result = classifier.assignClass();
        assertEquals("G1", ((HierarchyTree) result.getBestClass()).getName());
        assertTrue(0.2 > result.getPosteriorProb());

        //test the 8th sequence in G7, it is the same as the 9th sequence in G8
        // the classifier should randomly choose a genus (either G7 or G8) because the score will be tie
        parser.next();
        parser.next();
        pSeq = parser.next();
        assertEquals(pSeq.getSeqName(), "XPh2G7_child1");
        iterator = new GoodWordIterator(pSeq.getSeqString());
        int G7_count = 0;
        int G8_count = 0;
        for ( int run = 0; run < DecisionMaker.NUM_OF_RUNS; run++){
View Full Code Here
38
39
40
41
42
43
44
45
46
47
48
49
50
51
        TreeFactory factory = new TreeFactory(taxReader);

        File infileReader = new File(System.class.getResource("/test/classifier/testNBClassifierSet.fasta").getFile());

        LineageSequenceParser parser = new LineageSequenceParser(infileReader);

        while (parser.hasNext()) {
            factory.addSequence((LineageSequence) parser.next());
        }

        HierarchyTree root = factory.getRoot();
        assertEquals(root.getName(), "ROOT");
        assertEquals(root.getTotalSeqs(), 9);
View Full Code Here
TOP

Related Classes of edu.msu.cme.rdp.classifier.train.LineageSequenceParser

  • edu.msu.cme.rdp.classifier.train.validation.crossvalidate.CrossValidate
  • edu.msu.cme.rdp.classifier.train.validation.crossvalidate.RdmSelectTaxon
  • edu.msu.cme.rdp.classifier.train.validation.DecisionMakerTest
  • edu.msu.cme.rdp.classifier.train.validation.distance.CompareTrainingSets
  • edu.msu.cme.rdp.classifier.train.validation.distance.PairwiseSeqDistance
  • edu.msu.cme.rdp.classifier.train.validation.distance.TaxaSimilarityMain
  • edu.msu.cme.rdp.classifier.train.validation.HierarchyTreeTest
  • edu.msu.cme.rdp.classifier.train.validation.leaveoneout.LeaveOneOutTesterMain
  • edu.msu.cme.rdp.classifier.train.validation.movingwindow.MainMovingWindow
  • edu.msu.cme.rdp.classifier.train.validation.NBClassifierTest
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.