public static String toFasta(Sequence... sequences) {
StringBuilder sb = new StringBuilder();
for (Sequence sequence : sequences)
sb.append('>' + sequence.getName() + '\n' + sequence.seqString() + '\n');
return sb.toString();
}
/**
* Creates the Map required by the super class.
*
* @param s1
* @param s2
* @return
*/
private static Map<String, SymbolList> createHashMap(Sequence s1, Sequence s2) {
Map<String, SymbolList> m = new HashMap<String, SymbolList>();
m.put(s1.getName(), s1);
m.put(s2.getName(), s2);
return m;
}
/**
* This method provides a BLAST-like formated alignment from the given <code>String</code>s, in
* which the sequence coordinates and the information "Query" or "Sbjct", respectively is added to
* each line. Each line contains <code>width</code> sequence characters including the gap symbols
* plus the meta information. There is one white line between two pairs of sequences.
*
* @param width the number of symbols to be displayed per line.
* @return formated String.
* @throws BioException
*/
public String formatOutput(int width) throws BioException {
int i, j;
/*
* Highlights equal symbols within the alignment, String match/missmatch
* representation
*/
StringBuilder path = new StringBuilder();
for (i = 0; i < Math.min(queryEnd - queryStart, subjectEnd - subjectStart) + 1; i++) {
Symbol a = query.symbolAt(i + queryStart);
Symbol b = subject.symbolAt(i + subjectStart);
if (!a.equals(query.getAlphabet().getGapSymbol())
&& !b.equals(subject.getAlphabet().getGapSymbol())
&& ((subMatrix.getValueAt(a, b) >= 0) || a.equals(b))) {
path.append('|');
} else {
path.append(' ');
}
}
int maxLength = path.length();
/*
* Math.max(queryEnd - queryStart, subjectEnd - subjectStart) + 1;
*/
Formatter output = new Formatter();
output.format("%n Time (ms): %s%n", time);
output.format(" Length: %d%n", maxLength);
output.format(" Score: %d%n", getScore());
output.format(" Query: %s, Length: %d%n", query.getName(), query.length() - nGapsQ);
output.format(" Sbjct: %s, Length: %d%n", subject.getName(), subject.length() - nGapsS);
output.format(
" Identities: %d/%d, i.e., %d %% (query) and %d %% (sbjct)%n",
identicals,
maxLength,
Math.round(getPercentIdentityQuery()),
Math.round(getPercentIdentitySubject()));
output.format(
" Similars: %d/%d, i.e., %d %% (query) and %d %% (sbjct)%n",
similars,
maxLength,
Math.round(getPercentSimilarityQuery()),
Math.round(getPercentSimilaritySubject()));
output.format(
" No. gaps: %d (%d %%) in query and %d (%d %%) in sbjct%n",
nGapsQ, Math.round(getPercentGapsQuery()), nGapsS, Math.round(getPercentGapsTarget()));
int queryLPos = queryStart, queryRPos, pathLPos = 0, pathRPos;
int subjectLPos = subjectStart, subjectRPos;
int ql = queryLPos - 1, qr = queryLPos - 1, qgaps;
int sl = subjectLPos - 1, sr = subjectLPos - 1, sgaps;
int widthLeft = String.valueOf(Math.max(queryStart, queryEnd)).length();
int widthRight = String.valueOf(Math.max(queryEnd, subjectEnd)).length() + 1;
// Take width of the meta information into account.
width = Math.max(width - widthLeft - widthRight - 12, 2);
for (i = 1; i <= Math.ceil((double) maxLength / width); i++) {
// Query
queryRPos =
Math.min(
queryStart + i * width - 1,
Math.min(queryEnd, subjectEnd - subjectStart + queryStart));
qgaps = 0;
for (j = queryLPos; j <= queryRPos; j++) {
if (!query.symbolAt(j).equals(query.getAlphabet().getGapSymbol())) {
qr++;
} else {
qgaps++;
}
}
if (qgaps <= queryRPos - queryLPos) {
ql++;
}
output.format("%nQuery: %" + widthLeft + "d ", ql);
output.format("%s ", query.subStr(queryLPos, queryRPos));
output.format("%-" + widthRight + "d%n", qr);
queryLPos = queryRPos + 1;
ql = qr;
// Path
pathRPos = Math.min(i * width, path.length());
output.format(
"%-" + (widthLeft + 10) + "c%s",
Character.valueOf(' '),
path.substring(pathLPos, pathRPos));
pathLPos = pathRPos;
// Sbjct
subjectRPos =
Math.min(
subjectStart + i * width - 1,
Math.min(queryEnd - queryStart + subjectStart, subjectEnd));
sgaps = 0;
for (j = subjectLPos; j <= subjectRPos; j++) {
if (!subject.symbolAt(j).equals(subject.getAlphabet().getGapSymbol())) {
sr++;
} else {
sgaps++;
}
}
if (sgaps <= subjectRPos - subjectLPos) {
sl++;
}
output.format("%nSbjct: %" + widthLeft + "d ", sl);
output.format("%s ", subject.subStr(subjectLPos, subjectRPos));
output.format("%-" + widthRight + "d%n", sr);
subjectLPos = subjectRPos + 1;
sl = sr;
}
return output.toString();
}
public void main(String[] args) throws BioException, IOException {
if (count.length != args.length) {
System.err.println("The number of counts and output file arguments does not match!");
System.exit(1);
}
OutputStream[] outputStreams;
if ((args != null) && (args.length > 0)) {
outputStreams = new OutputStream[args.length];
for (int i = 0; i < args.length; i++) {
outputStreams[i] = new BufferedOutputStream(new FileOutputStream(args[i]));
}
} else {
outputStreams = new OutputStream[] {System.out};
}
RichSequenceIterator seqIterator =
RichSequence.IOTools.readFastaDNA(new BufferedReader(new FileReader(seqFile)), null);
List<Sequence> seqs = new ArrayList<Sequence>();
while (seqIterator.hasNext()) {
seqs.add(seqIterator.nextSequence());
}
List<List<Sequence>> chosenSeqs = new ArrayList<List<Sequence>>();
if (!sampleWithReplacement) {
for (int c : count) {
List<Sequence> cseqs = new ArrayList<Sequence>();
chosenSeqs.add(cseqs);
while (c > 0) {
int randSeqIndex = random.nextInt(seqs.size());
cseqs.add(seqs.remove(randSeqIndex));
c--;
}
}
} else if (sampleWithReplacement || (length > 0)) {
/*
* if you want to sample from sequences with replacement
* or if the wanted length is specified
*/
for (int c : count) {
int i = 0;
List<Sequence> cseqs = new ArrayList<Sequence>();
chosenSeqs.add(cseqs);
while (c > 0) {
Sequence randomSeq = seqs.get(random.nextInt(seqs.size()));
if (length > 0) {
int startPos = random.nextInt(1 + randomSeq.length() - length);
cseqs.add(
new SimpleSequence(
randomSeq.subList(startPos, startPos + length),
null,
randomSeq.getName() + "_" + i++,
Annotation.EMPTY_ANNOTATION));
} else {
cseqs.add(randomSeq);
}
c--;
}
}
}
int i = 0;
for (List<Sequence> seqList : chosenSeqs) {
for (Sequence seq : seqList) {
Sequence s;
if (uniqueNames) {
s =
new SimpleSequence(
seq.subList(1, seq.length()),
null,
seq.getName() + "_" + i,
Annotation.EMPTY_ANNOTATION);
} else {
s = seq;
}
RichSequence.IOTools.writeFasta(outputStreams[i], s, null);
outputStreams[i].flush();
}
i++;
}
seqs = null;
}