private final int removeDiscarded(ArrayList<Read> buffer1, ArrayList<Read> buffer2) {
int removed = 0;
if (buffer2 == null) {
for (int i = 0; i < buffer1.size(); i++) {
Read a = buffer1.get(i);
if (a.discarded()) {
buffer1.set(i, null);
removed++;
}
}
} else {
for (int i = 0; i < buffer1.size(); i++) {
Read a = buffer1.get(i);
Read b = buffer2.get(i);
if (a.discarded() || b.discarded()) {
buffer1.set(i, null);
buffer2.set(i, null);
removed++;
}
}
}
if (removed > 0) {
Tools.condenseStrict(buffer1);
if (buffer2 != null) {
Tools.condenseStrict(buffer2);
}
}
return removed;
}
@Override
boolean processReadPair(Read r1, Read r2) {
assert (r2 == null);
final byte[] quals = r1.quality, bases = r1.bases;
final byte[] match =
(r1.match == null ? null : !r1.shortmatch() ? r1.match : Read.toLongMatchString(r1.match));
if (match == null || quals == null || bases == null) {
return false;
}
int subs = 0;
int indels = 0;
for (int qpos = 0, mpos = 0, last = quals.length - 1; mpos < match.length; mpos++) {
final byte m = match[mpos];
final byte mprev = match[Tools.max(mpos - 1, 0)];
final byte mnext = match[Tools.min(mpos + 1, match.length - 1)];
final byte q1 = quals[qpos];
final byte b2 = bases[qpos];
int sub = 0, indel = 0;
if (m == 'S') {
sub = 1;
} else if (m == 'I') {
indel = 1;
} else if (m == 'm') {
if (mprev == 'D' || mnext == 'D') {
indel = 1;
}
} else if (m == 'D') {
// do nothing
} else if (m == 'C') {
// do nothing
} else {
throw new RuntimeException(
"Bad symbol m='"
+ ((char) m)
+ "'\n"
+ new String(match)
+ "\n"
+ new String(bases)
+ "\n");
}
subs += sub;
indels += indel;
if (q1 >= minq && q1 <= maxq) {
if (sub > 0 || (indel > 0 && countIndels)) {
return true;
}
}
if (m != 'D') {
qpos++;
}
}
return keepPerfect && subs == 0 && indels == 0;
}
public Read consensus() { // TODO: Return single read if only 1.
final int[][] counts = baseCounts();
final int width = counts[0].length;
byte[] bases = new byte[width], quals = new byte[width];
for (int i = 0; i < width; i++) {
int x = getConsensus(counts, i);
if (x < 0) {
// System.err.println("q="+0+", x="+x+"; A="+counts[0][i]+", C="+counts[1][i]+",
// G="+counts[2][i]+", T="+counts[3][i]);
bases[i] = 'N';
quals[i] = 0;
} else {
long q = 2 * counts[x][i] - counts[0][i] - counts[1][i] - counts[2][i] - counts[3][i];
// System.err.println("q="+q+", x="+x+"; A="+counts[0][i]+", C="+counts[1][i]+",
// G="+counts[2][i]+", T="+counts[3][i]);
bases[i] = AminoAcid.numberToBase[x];
quals[i] = (byte) Tools.mid(0, q, 50);
}
}
Read leftmost = this.get(0);
Read r = new Read(bases, quals, 0, leftmost.id);
// TODO: Attach the long pair, and make sure the kmer location is correct.
// assert(false) : "\n"+r.toFastq()+"\nCheck kmer location.";
// assert(size()==1) :
// "\n"+r.toFastq()+"\n"+get(0).toFastq()+"\n"+get(size()-1).toFastq()+"\n";
return r;
}
@Override
public boolean parseArgument(String arg, String a, String b) {
// System.err.println("Calling parseArgument("+arg+","+a+","+b+")");
if (a.equals("minq")) {
minq = (int) Tools.parseKMG(b);
return true;
} else if (a.equals("maxq")) {
maxq = (int) Tools.parseKMG(b);
return true;
} else if (a.equals("keepperfect")) {
keepPerfect = Tools.parseBoolean(b);
return true;
} else if (a.equals("countindels")) {
countIndels = Tools.parseBoolean(b);
return true;
}
// There was no match to the argument
return false;
}
private final void pair(ArrayList<Read> buffer1, ArrayList<Read> buffer2) {
final int len1 = buffer1.size(), len2 = buffer2.size();
assert (ALLOW_UNEQUAL_LENGTHS || len1 == len2)
: "\nThere appear to be different numbers of reads in the paired input files."
+ "\nThe pairing may have been corrupted by an upstream process. It may be fixable by running repair.sh.";
final int lim = Tools.min(len1, len2);
for (int i = 0; i < lim; i++) {
Read a = buffer1.get(i);
Read b = buffer2.get(i);
assert (a.numericID == b.numericID)
: "\n"
+ a.numericID
+ ", "
+ b.numericID
+ "\n"
+ a.toText(false)
+ "\n"
+ b.toText(false)
+ "\n";
assert (a.mate == null)
: "Please set interleaved=false when using dual input files.\n"
+ a.id
+ "\n"
+ a.mate.id
+ "\n"
+ b.id
+ "\n"
+ producer1
+ "\n"
+ producer2;
assert (b.mate == null) : "Please set interleaved=false when using dual input files.";
a.mate = b;
b.mate = a;
assert (a.pairnum() == 0);
b.setPairnum(1);
// assert(a.pairnum()!=b.pairnum());
}
if (len1 > len2) {
// do nothing;
} else if (len2 > len1) {
for (int i = lim; i < len2; i++) {
Read b = buffer2.get(i);
b.setPairnum(0);
buffer1.add(b);
}
}
}
/** This will create a count consensus of the bases at each position in the cluster. */
public int[][] baseCounts() {
int maxLeft = -1, maxRight = -1;
for (Read r : this) {
long[] obj = (long[]) r.obj;
int pos = (int) obj[1];
maxLeft = Tools.max(maxLeft, pos);
maxRight = Tools.max(maxRight, r.length() - pos);
}
final int width = maxLeft + maxRight;
// assert(size()==1) : "\nleft="+maxLeft+", right="+maxRight+", width="+width+",
// "+k+"\n"+get(0).toFastq()+"\n"+get(size()-1).toFastq();
// System.err.println("\n\n");
final int[][] counts = new int[4][width];
for (Read r : this) {
long[] obj = (long[]) r.obj;
int pos = (int) obj[1];
byte[] bases = r.bases, quals = r.quality;
// System.err.println("pos="+pos+", maxLeft="+maxLeft);
for (int cloc = 0, rloc = maxLeft - pos; cloc < bases.length; cloc++, rloc++) {
// System.err.println("cloc="+cloc+"/"+bases.length+", rloc="+rloc+"/"+width);
int x = AminoAcid.baseToNumber[bases[cloc]];
if (x > -1) {
int q = (quals == null ? 20 : quals[cloc]);
counts[x][rloc] += q;
}
}
}
// if(size()>0){//Looks correct.
// System.err.println(Arrays.toString(counts[0]));
// System.err.println(Arrays.toString(counts[1]));
// System.err.println(Arrays.toString(counts[2]));
// System.err.println(Arrays.toString(counts[3]));
// }
return counts;
}
public void fill(Read r, long[] kmers) {
final byte[] bases = r.bases;
long kmer = 0;
long rkmer = 0;
int len = 0;
if (bases == null || bases.length < k) {
return;
}
long[] mods = local2.get();
if (mods == null) {
mods = new long[comparisons];
local2.set(mods);
}
Arrays.fill(mods, -1);
for (int i = 0; i < bases.length; i++) {
byte b = bases[i];
long x = Dedupe.baseToNumber[b];
long x2 = Dedupe.baseToComplementNumber[b];
kmer = ((kmer << 2) | x) & mask;
rkmer = (rkmer >>> 2) | (x2 << shift2);
if (b == 'N') {
len = 0;
} else {
len++;
}
if (len >= k) {
final long kmax = Tools.max(kmer, rkmer);
for (int j = 0; j < comparisons; j++) {
final long div = divisors[j];
final long mod = kmax % div;
if (mod > mods[j]) {
mods[j] = mod;
kmers[j] = kmax;
}
}
}
}
}
public static void main(String[] args) {
Timer t = new Timer();
String in = args[0];
String outF = args.length > 1 ? args[1] : null;
String outR = args.length > 2 ? args[2] : null;
String outU = args.length > 3 ? args[3] : null;
if (args.length > 4) {
if (args[4].equalsIgnoreCase("header")) {
includeHeader = true;
}
}
ByteFile tf = ByteFile.makeByteFile(in, true, false);
Tools.testForDuplicateFiles(true, in, outF, outR, outU);
Tools.testOutputFiles(true, false, false, outF, outR, outU);
final ByteStreamWriter fStream, rStream, uStream;
fStream = (outF == null ? null : new ByteStreamWriter(outF, true, false, true));
rStream = (outR == null ? null : new ByteStreamWriter(outR, true, false, true));
uStream = (outU == null ? null : new ByteStreamWriter(outU, true, false, true));
if (fStream != null) {
fStream.start();
}
if (rStream != null) {
rStream.start();
}
if (uStream != null) {
uStream.start();
}
long plus = 0;
long minus = 0;
long other = 0;
byte[] s = null;
for (s = tf.nextLine(); s != null; s = tf.nextLine()) {
if (s.length > 0) {
byte c = s[0];
if (c == '@') {
if (includeHeader) {
if (fStream != null) {
fStream.println(s);
}
if (rStream != null) {
rStream.println(s);
}
if (uStream != null) {
uStream.println(s);
}
}
} else {
int flag = SamLine.parseFlagOnly(s);
if (SamLine.mapped(flag)) {
if (SamLine.strand(flag) == 0) {
if (fStream != null) {
fStream.println(s);
}
plus++;
} else {
if (rStream != null) {
rStream.println(s);
}
minus++;
}
} else {
if (uStream != null) {
uStream.println(s);
}
other++;
}
}
}
}
tf.close();
if (fStream != null) {
fStream.poisonAndWait();
}
if (rStream != null) {
rStream.poisonAndWait();
}
if (uStream != null) {
uStream.poisonAndWait();
}
System.err.println("Total reads: \t" + (plus + minus + other));
System.err.println("Plus reads: \t" + (plus));
System.err.println("Minus reads: \t" + (minus));
System.err.println("Unmapped reads:\t" + (other));
t.stop();
System.err.println("Time: \t" + t);
}
public static void main(String[] args) {
System.err.println(
"Executing "
+ (new Object() {}.getClass().getEnclosingClass().getName())
+ " "
+ Arrays.toString(args)
+ "\n");
Timer t = new Timer();
String inPattern = args[0];
int minChrom = -1;
int maxChrom = -1;
int outgenome = -1;
Data.GENOME_BUILD = -1;
String name = null;
for (int i = 1; i < args.length; i++) {
final String arg = args[i].toLowerCase();
String[] split = arg.split("=");
String a = split[0];
String b = (split.length > 1 ? split[1] : null);
if (a.equals("ingenome")) {
Data.setGenome(Integer.parseInt(b));
if (minChrom == -1) {
minChrom = 1;
}
if (maxChrom == -1) {
maxChrom = Data.numChroms;
}
} else if (a.equals("outgenome")) {
outgenome = Integer.parseInt(b);
} else if (a.equals("minchrom")) {
minChrom = Integer.parseInt(b);
} else if (a.equals("maxchrom")) {
maxChrom = Integer.parseInt(b);
} else if (a.equals("threads") || a.equals("t")) {
THREADS = Integer.parseInt(b);
} else if (a.equals("nblocksize")) {
N_BLOCK_SIZE = Integer.parseInt(b);
} else if (a.equals("nblocktrigger")) {
N_BLOCK_TRIGGER = Integer.parseInt(b);
} else if (a.equals("staynearref")) {
STAY_NEAR_REF = Tools.parseBoolean(b);
} else if (a.equals("append") || a.equals("app")) {
append = ReadStats.append = Tools.parseBoolean(b);
} else if (a.equals("overwrite") || a.equals("ow")) {
overwrite = Tools.parseBoolean(b);
} else if (a.startsWith("regen")) {
REGEN_N_BLOCKS = Tools.parseBoolean(b);
} else if (a.startsWith("name=")) {
REGEN_N_BLOCKS = Tools.parseBoolean(b);
} else {
System.err.println("Unknown argument " + arg);
}
}
assert (Data.GENOME_BUILD > -1);
assert (outgenome > -1);
// assert(Data.GENOME_BUILD!=outgenome);
if (Data.GENOME_BUILD == outgenome) {
System.out.println("Warning! Overwriting input genome " + outgenome);
}
String fname = Data.chromFname(minChrom, outgenome);
File f = new File(fname.substring(0, fname.lastIndexOf('/')));
// assert(false) : f.getAbsolutePath();
if (!f.exists()) {
f.mkdirs();
}
for (int chrom = minChrom; chrom <= maxChrom; chrom++) {
String outName = Data.chromFname(chrom, outgenome);
assert (overwrite || !new File(outName).exists())
: "Destination " + outName + " already exists.";
// assert(false) : inPattern+", "+outName;
process(inPattern.replaceFirst("#", "" + chrom), outName, chrom);
}
FastaToChromArrays2.writeInfo(
outgenome,
maxChrom,
(name == null ? Data.name : name),
"" + Data.GENOME_BUILD + "_plus_variations",
false,
false);
t.stop();
{
String path = IndexMaker4.fname(1, 1, 12, 1);
int lastSlash = path.lastIndexOf('/');
path = path.substring(0, lastSlash);
File dir = new File(path);
if (dir.exists()) {
System.out.println("Deleting old index for " + outgenome);
for (File f2 : dir.listFiles()) {
if (f2.isFile() && (f2.getName().contains(".int2d") || f2.getName().endsWith(".txt"))) {
f2.delete();
}
}
}
}
// System.out.println("Vars in: \t"+VARS_IN);
// System.out.println("Vars out:\t"+VARS_OUT);
System.out.println();
System.out.println("Time: \t" + t);
}
/**
* @param replaceFirst
* @param chromFname
* @param chrom
*/
public static void process(String inVarsName, String outChromName, int chrom) {
ArrayList<Varlet> vars = Varlet.fromTextFile(inVarsName);
ChromosomeArray cha = Data.getChromosome(chrom);
ChromosomeArray chb = new ChromosomeArray(chrom, Gene.PLUS);
// Next location to read in a
int aloc = 0;
// Next location to set in b
int bloc = 0;
for (int i = 0; i < vars.size(); i++) {
Varlet v = vars.get(i);
assert (v.beginLoc >= aloc)
: i + "\n" + vars.get(i - 1) + "\n" + v + "\n"; // Overlapping variations
while (v.beginLoc < aloc) { // skip it, for now.
System.err.print("e");
i++;
if (i >= vars.size()) {
break;
}
v = vars.get(i);
}
if (STAY_NEAR_REF && Tools.absdif(aloc, bloc) >= REF_LIMIT) {
int dif = v.lengthDif();
if (aloc < bloc) { // skip insertions
while (dif > 0) {
// System.err.print("i");
i++;
if (i >= vars.size()) {
break;
}
v = vars.get(i);
dif = v.lengthDif();
}
} else { // skip deletions
while (dif < 0) {
// System.err.print("d");
i++;
if (i >= vars.size()) {
break;
}
v = vars.get(i);
dif = v.lengthDif();
}
}
}
// Advance to variation's beginning
while (aloc < v.beginLoc) {
byte b = cha.get(aloc);
chb.set(bloc, b);
aloc++;
bloc++;
}
// Apply variation
if (v.varType == Variation.SNP) {
String call = v.call;
String ref = v.ref;
if (ref != null && ref.equals("=")) {
ref = null;
}
for (int j = 0; j < call.length(); j++) {
char c = call.charAt(j);
if (ref != null) {
assert (ref.charAt(j) == cha.get(aloc)) : "\n" + i + ", " + v;
}
chb.set(bloc, c);
aloc++;
bloc++;
}
} else if (v.varType == Variation.DELINS) {
String call = v.call;
for (int j = 0; j < call.length(); j++) {
char c = call.charAt(j);
chb.set(bloc, c);
bloc++;
}
aloc += v.lengthRef();
} else if (v.varType == Variation.NOCALL) {
// Do nothing. But, it should have been removed already.
if (!foundNocall) {
System.err.println("*** Warning - found a nocall in input variations ***");
foundNocall = true;
}
} else if (v.varType == Variation.NOREF) {
String call = v.call;
for (int j = 0; j < call.length(); j++) {
char c = call.charAt(j);
assert (cha.get(aloc) == 'N') : cha.get(aloc);
chb.set(bloc, c);
aloc++;
bloc++;
}
} else if (v.varType == Variation.INS) {
String call = v.call;
for (int j = 0; j < call.length(); j++) {
char c = call.charAt(j);
chb.set(bloc, c);
bloc++;
}
} else if (v.varType == Variation.DEL) {
int len = v.lengthRef();
assert (len > 0);
aloc += len;
}
}
// Finish writing array
while (aloc < cha.array.length || aloc <= cha.maxIndex) {
byte c = cha.get(aloc);
chb.set(bloc, c);
aloc++;
bloc++;
}
System.out.println("Length Shift for chr" + chrom + ": \t" + (bloc - aloc));
Data.unload(chrom, true);
cha = null;
if (REGEN_N_BLOCKS) {
chb = regenNBlocks(chb, N_BLOCK_SIZE, N_BLOCK_TRIGGER, N_BLOCK_END_SIZE);
}
chb.resize(chb.maxIndex + 1);
// Can't do this because it is read later
// if(THREADS==1){ReadWrite.writeObjectInThread(cac, outChromName);}
// else{ReadWrite.write(cac, outChromName);}
ReadWrite.write(chb, outChromName, false);
}
public static void main(String[] args) {
String in1 = args[0];
String in2 =
(args.length < 2 || args[1].equalsIgnoreCase("null") || args[1].contains("=")
? null
: args[1]);
if (in2 != null) {
assert (!in1.equalsIgnoreCase(in2));
FASTQ.TEST_INTERLEAVED = false;
} else {
FASTQ.TEST_INTERLEAVED = true;
FASTQ.FORCE_INTERLEAVED = true;
}
long maxReads = -1;
for (int i = 1; i < args.length; i++) {
final String arg = args[i];
final String[] split = arg.split("=");
String a = split[0].toLowerCase();
String b = (split.length > 1 ? split[1] : "true");
if (Parser.isJavaFlag(arg)) {
// jvm argument; do nothing
} else if (Parser.parseZip(arg, a, b)) {
// do nothing
} else if (Parser.parseQuality(arg, a, b)) {
// do nothing
} else if (Parser.parseFasta(arg, a, b)) {
// do nothing
} else if (a.equals("reads") || a.startsWith("maxreads")) {
maxReads = Tools.parseKMG(b);
} else {
throw new RuntimeException("Unknown parameter " + args[i]);
}
}
Parser.processQuality();
assert (FastaReadInputStream.settingsOK());
Timer t = new Timer();
ConcurrentReadInputStream cris = getReadInputStream(maxReads, false, true, in1, in2);
System.out.println("Fetched " + cris.getClass().getName());
{
Object[] p = cris.producers();
// while(p[0]==null){
// p=cris.producers();
// }
System.out.print("Producers: ");
String comma = "";
for (Object o : p) {
System.out.print(comma + (o == null ? "null" : o.getClass().getName()));
comma = ", ";
}
System.out.println();
}
boolean paired = cris.paired();
System.out.println("paired=" + paired);
cris.start(); // 4567
ListNum<Read> ln = cris.nextList();
ArrayList<Read> reads = (ln != null ? ln.list : null);
if (reads != null && !reads.isEmpty()) {
Read r = reads.get(0);
assert ((r.mate != null) == paired);
}
long readCount = 0;
long baseCount = 0;
while (reads != null && reads.size() > 0) {
for (Read r : reads) {
Read r2 = r.mate;
if (r != null) {
readCount++;
if (r.bases != null) {
baseCount += r.length();
}
}
if (r2 != null) {
readCount++;
if (r2.bases != null) {
baseCount += r2.length();
}
}
}
cris.returnList(ln.id, ln.list.isEmpty());
// System.err.println("fetching list");
ln = cris.nextList();
reads = (ln != null ? ln.list : null);
// System.out.println("reads: "+(reads==null ? "null" : reads.size()));
}
System.err.println("Finished reading");
cris.returnList(ln.id, ln.list.isEmpty());
cris.close();
t.stop();
System.out.println("Reads: \t" + readCount);
System.out.println("Bases: \t" + baseCount);
System.out.println("Avg Length: \t" + String.format("%.2f", baseCount * 1.0 / readCount));
System.out.println("Time: \t" + t);
}
