public MultipleAlignment deepClone(int ntxinc) {
MultipleAlignment ma = new MultipleAlignment(nTaxon, seqLen, nTaxon + ntxinc);
System.arraycopy(names, 0, ma.names, 0, names.length);
System.arraycopy(data, 0, ma.data, 0, data.length);
ma.buildNameIndex();
return ma;
}
public static MultipleAlignment loadPhylip(Reader rarg) {
try {
BufferedReader r = new BufferedReader(rarg);
String header = r.readLine();
StringTokenizer st = new StringTokenizer(header);
int nTaxon = Integer.parseInt(st.nextToken());
int seqLen = Integer.parseInt(st.nextToken());
MultipleAlignment ma = new MultipleAlignment(nTaxon, seqLen);
for (int i = 0; i < nTaxon; i++) {
String line = r.readLine();
if (line == null) {
throw new RuntimeException("cannot read next line in " + "input file");
}
st = new StringTokenizer(line);
String name = st.nextToken();
String data = st.nextToken();
if (data.length() != seqLen) {
throw new RuntimeException("wrong sequence length: " + data.length() + " vs " + seqLen);
}
ma.names[i] = name;
ma.data[i] = data.toUpperCase();
}
ma.buildNameIndex();
r.close();
return ma;
} catch (IOException ex) {
Logger.getLogger(MultipleAlignment.class.getName()).log(Level.SEVERE, null, ex);
throw new RuntimeException("bailing out");
}
}
public static void main(String args[]) {
if (false) {
// MultipleAlignment ma = MultipleAlignment.loadPhylipShitty(new File(
// "/space/raxml/VINCENT/DATA/500"));
long time1 = System.currentTimeMillis();
MultipleAlignment mar =
MultipleAlignment.loadPhylip(new File("/space/raxml/VINCENT/DATA/150"));
MultipleAlignment ma =
MultipleAlignment.loadPhylipShitty(new File("/space/raxml/VINCENT/DATA/150"));
System.out.printf(
"parse done: %d %s\n", System.currentTimeMillis() - time1, compare(ma, mar));
// ma.print();
} else {
long time1 = System.currentTimeMillis();
MultipleAlignment ma =
MultipleAlignment.loadPhylipShitty(new File("/space/raxml/VINCENT/DATA/500"));
System.out.printf("parse done: %d\n", System.currentTimeMillis() - time1);
}
}
public static MultipleAlignment loadPhylipShitty(File file) {
try {
BufferedReader r = new BufferedReader(new FileReader(file));
String header = r.readLine();
StringTokenizer st = new StringTokenizer(header);
final int nTaxon = Integer.parseInt(st.nextToken());
final int seqLen = Integer.parseInt(st.nextToken());
MultipleAlignment ma = new MultipleAlignment(nTaxon, seqLen);
ma.nameMap = new HashMap<String, Integer>();
String line = null;
boolean keepline = false;
mainloop:
while (true) {
for (int i = 0; i < nTaxon; i++) {
if (!keepline) {
line = r.readLine();
}
keepline = false;
if (line == null) {
break mainloop;
}
String nd[] = readLine(line);
String name = nd[0];
String data = nd[1];
if (name != null) {
assert (!ma.nameMap.containsKey(name));
ma.names[i] = name;
ma.data[i] = data;
ma.nameMap.put(name, i);
} else {
ma.data[i] += data;
}
}
//
// look if there are non-empty lines following the last block.
// keepline is used to implement a crude form of 'unreadline'.
//
keepline = false;
outer:
while ((line = r.readLine()) != null) {
// System.out.printf( "look: '%s'\n", line );
for (int i = 0; i < line.length(); i++) {
if (!Character.isWhitespace(line.charAt(i))) {
keepline = true;
break outer;
}
}
}
if (!keepline) {
break;
}
}
// sanity check on sequence length
for (int i = 0; i < nTaxon; i++) {
if (ma.nameMap.get(ma.names[i]) != i) {
throw new RuntimeException("quirk in name index");
}
if (ma.data[i].length() != seqLen) {
throw new RuntimeException("wrong sequence length ");
}
}
r.close();
return ma;
} catch (IOException ex) {
Logger.getLogger(MultipleAlignment.class.getName()).log(Level.SEVERE, null, ex);
throw new RuntimeException("bailing out");
}
}