public static void main(String[] args) throws IOException {
if (args.length != 2) {
System.err.println("USAGE: AllTranslate <in_nucl_file> <out_file>");
System.exit(1);
}
SeqReader reader = new SequenceReader(new File(args[0]));
FastaWriter writer = new FastaWriter(new PrintStream(args[1]));
Sequence seq;
ProteinUtils protUtils = ProteinUtils.getInstance();
while ((seq = reader.readNextSequence()) != null) {
for (Sequence protSeq : protUtils.allTranslate(seq)) {
writer.writeSeq(protSeq);
}
}
writer.close();
}
private static void convertToFasta(File sffFile, File workDir, String outfilePrefix)
throws IOException {
SFFCore core = new SFFCore(sffFile);
FastaWriter fastaWriter = new FastaWriter(new File(workDir, outfilePrefix + ".fasta"));
FastaWriter qualWriter = new FastaWriter(new File(workDir, outfilePrefix + ".qual"));
QSequence seq;
while ((seq = core.readNextSeq()) != null) {
fastaWriter.writeSeq(seq);
qualWriter.writeSeq(seq.getSeqName(), "", SeqUtils.translateQualString(seq.getQuality()));
}
qualWriter.close();
fastaWriter.close();
core.close();
}
public static void main(String[] args) throws Exception {
if ((args.length < 6 && args.length != 1) || args.length > 13) {
printUsage(true);
System.exit(1);
}
// check for optional arguments
boolean normalized = true;
int heuristicPruning = HMMGraphSearch.PRUNE_NODE;
String hweightstr = "static";
double weight = 1.0;
int optCount = 0;
for (int i = 0; i < 6; ++i) {
if (args[i].equals("-u")) {
normalized = false;
optCount++;
} else if (args[i].equals("-p")) {
heuristicPruning = Integer.parseInt(args[i + 1]);
i++;
optCount += 2;
} else if (args[i].equals("-w")) {
weight = Double.parseDouble(args[i + 1]);
i++;
optCount += 2;
} else if (args[i].equals("-m")) {
hweightstr = args[i + 1];
i++;
optCount += 2;
} else if (args[i].equals("-h")) {
printUsage(true);
System.exit(1);
}
}
if (optCount > 0) {
args = Arrays.copyOfRange(args, optCount, args.length);
}
int k = Integer.valueOf(args[0]);
long timeLimit = Long.valueOf(args[1]);
File bloomFile = new File(args[2]);
File forHMMFile = new File(args[3]);
File revHMMFile = new File(args[4]);
File kmersFile = new File(args[5]);
File nuclOutFile = new File(kmersFile.getName() + "_nucl.fasta");
File protOutFile = new File(kmersFile.getName() + "_prot.fasta");
HeuristicWeight hweight;
if (hweightstr.equals("static")) {
hweight = new StaticHeuristicWeight(weight);
} else if (hweightstr.equals("dynamic")) {
hweight = new DynamicHeuristicWeight(weight);
} else if (hweightstr.equals("revised_dynamic")) {
hweight = new RevisedDynamicHeuristicWeight(weight);
} else {
throw new RuntimeException("Invalid argument for heuristic weight: " + hweightstr);
}
ProfileHMM forHMM;
ProfileHMM revHMM;
if (normalized) {
forHMM = HMMER3bParser.readModel(forHMMFile);
revHMM = HMMER3bParser.readModel(revHMMFile);
} else {
forHMM = HMMER3bParser.readUnnormalized(forHMMFile);
revHMM = HMMER3bParser.readUnnormalized(revHMMFile);
}
FastaWriter nuclOut = new FastaWriter(nuclOutFile);
FastaWriter protOut = null;
boolean isProt = forHMM.getAlphabet() == SequenceType.Protein;
if (isProt) {
protOut = new FastaWriter(protOutFile);
}
int kmerCount = 0;
int contigCount = 0;
long startTime;
startTime = System.currentTimeMillis();
ObjectInputStream ois =
new ObjectInputStream(new BufferedInputStream(new FileInputStream(bloomFile)));
BloomFilter bloom = (BloomFilter) ois.readObject();
ois.close();
System.err.println(
"Bloom filter loaded in " + (System.currentTimeMillis() - startTime) + " ms");
System.err.println("Starting hmmgs search at " + new Date());
System.err.println("* Kmer file: " + kmersFile);
System.err.println("* Bloom file: " + bloomFile);
System.err.println("* Forward hmm file: " + forHMMFile);
System.err.println("* Reverse hmm file: " + revHMMFile);
System.err.println("* Searching prot?: " + isProt);
System.err.println("* # paths: " + k);
System.err.println("* Nucl contigs out file " + nuclOutFile);
System.err.println("* Prot contigs out file " + protOutFile);
System.err.println("* heuristicPruning " + heuristicPruning);
// these are experimental options, not used
// System.err.println("* HeuristicWeightMethod " + hweightstr);
// System.err.println("* HeuristicWeight " + weight);
startTime = System.currentTimeMillis();
HMMBloomSearch.printHeader(System.out, isProt);
KmerStart line;
KmerStartsReader reader = new KmerStartsReader(kmersFile);
HMMGraphSearch search = new HMMGraphSearch(k, heuristicPruning);
try {
while ((line = reader.readNext()) != null) {
search.setHWeight(hweight);
kmerCount++;
if (line.getMpos() == 0) {
System.err.println("Skipping line " + line.getKmer());
continue;
}
TimeStamppedFutureTask future =
new TimeStamppedFutureTask(
new TimeLimitedSearchThread(
search,
new SearchTarget(
line.getGeneName(),
line.getQueryId(),
line.getRefId(),
line.getNuclKmer(),
0,
line.getMpos() - 1,
forHMM,
revHMM,
bloom)));
Thread t = new Thread(future);
t.setDaemon(true);
t.setPriority(Thread.MAX_PRIORITY);
t.start();
try {
List<SearchResult> searchResults = future.get(timeLimit, TimeUnit.SECONDS);
for (SearchResult result : searchResults) {
String seqid = "contig_" + (contigCount++);
HMMBloomSearch.printResult(seqid, isProt, result, System.out);
nuclOut.writeSeq(seqid, result.getNuclSeq());
if (isProt) {
protOut.writeSeq(seqid, result.getProtSeq());
}
}
} catch (TimeoutException e) {
System.out.println(
"-\t" + future.getStartingWord() + (isProt ? "\t-" : "") + "\t-\t-\t-\t-");
future.cancel(true);
} catch (Exception e) {
System.out.println(
"-\t" + future.getStartingWord() + (isProt ? "\t-" : "") + "\t-\t-\t-\t-");
e.printStackTrace();
if (e.getCause() != null) {
e.getCause().printStackTrace();
}
future.cancel(true);
}
}
System.err.println(
"Read in "
+ kmerCount
+ " kmers and created "
+ contigCount
+ " contigs in "
+ (System.currentTimeMillis() - startTime) / 1000f
+ " seconds");
} finally {
nuclOut.close();
if (isProt) {
protOut.close();
}
System.out.close();
}
}