1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
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
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
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
|
package pp2.prediction.knn;
import java.io.BufferedReader;
import java.io.BufferedWriter;
import java.io.IOException;
import pp2.go.Relations;
import pp2.tools.Tools;
/**
* performs all operations necessary to predict a list of targets
*
* @author Thomas Hopf
*
*/
public class KNNKernel {
// holds all relations in the GO ontology tree
private Relations relations;
public KNNKernel(String closureFile, String termFile, boolean noCellularComponent)
{
relations = new Relations(closureFile, termFile, noCellularComponent);
}
/**
* fetches the next target with all BLAST hits from the reader
* @param reader
* @return
*/
private PredictionTarget getNextTarget(BufferedReader reader, boolean eliminateDuplicateHits)
{
PredictionTarget nextTarget = new PredictionTarget();
try {
String line = null;
while((line = reader.readLine()) != null)
{
// return current target when we hit the "end-of-target"-line "//"
if(line.equals("//"))
return nextTarget;
// get ID of current target
if(line.startsWith(">"))
nextTarget.setTargetID(line.substring(1));
// if line starts with "#", this is one of the BLAST hits in the database
if(line.startsWith("#"))
nextTarget.addBlastHit(line.substring(1), eliminateDuplicateHits);
}
} catch(Exception e)
{
e.printStackTrace();
}
return null;
}
/**
* carries out the prediction of all targets in file containing multiple targets
* @param targetFile
*/
public void predictTargetList(String targetFile, String outFile, boolean eliminateDuplicateHits, boolean useEValueThreshold, double eValueTreshold, int k, String distributionFile)
{
try {
BufferedReader in = Tools.openFile(targetFile);
BufferedWriter naiveWriter = Tools.writeToFile(outFile + ".naive");
BufferedWriter leaveWriter = Tools.writeToFile(outFile + ".leaves");
// read the distribution of global quality scores
double[] distribution = null;
if(distributionFile != null)
{
distribution = new double[100];
int i = 0;
BufferedReader distReader = Tools.openFile(distributionFile);
String line = null;
while((line = distReader.readLine()) != null)
{
distribution[i] = Double.parseDouble(line);
i++;
}
if(i != 100)
System.err.println("warning: distribution file does not contain exactly 100 quantile lines.");
distReader.close();
}
// the current target for which to perform a prediction
PredictionTarget target = null;
int numPredicted = 0;
// iterate over all targets
while((target = getNextTarget(in, eliminateDuplicateHits)) != null)
{
// skip double // error in blast output file
if(target.getTargetID() == null)
continue;
//System.err.println(">" + target.getTargetID());
try {
target.predict(relations, useEValueThreshold, eValueTreshold, k, naiveWriter, leaveWriter, distribution);
} catch(Exception e)
{
System.err.println("an error occured while predicting target " + target.getTargetID());
e.printStackTrace();
}
numPredicted++;
}
System.out.println(" predicted " + numPredicted + " sequences.");
in.close();
naiveWriter.close();
leaveWriter.close();
} catch(IOException e)
{
e.printStackTrace();
}
}
}
|
