Mercurial > repos > rliterman > csp2

annotate CSP2/CSP2_env/env-d9b9114564458d9d-741b3de822f2aaca6c6caa4325c4afce/opt/bbmap-39.01-1/current/bloom/KmerCount4.java @ 68:5028fdace37b

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload commit 2e9511a184a1ca667c7be0c6321a36dc4e3d116d
author jpayne
date Tue, 18 Mar 2025 16:23:26 -0400
parents
children
rev   line source
jpayne@68 1 package bloom;
jpayne@68 2
jpayne@68 3 import java.util.ArrayList;
jpayne@68 4 import java.util.Locale;
jpayne@68 5
jpayne@68 6 import dna.AminoAcid;
jpayne@68 7 import fileIO.FileFormat;
jpayne@68 8 import shared.Timer;
jpayne@68 9 import stream.ConcurrentReadInputStream;
jpayne@68 10 import stream.FastaReadInputStream;
jpayne@68 11 import stream.Read;
jpayne@68 12 import structures.ListNum;
jpayne@68 13
jpayne@68 14 /**
jpayne@68 15 * @author Brian Bushnell
jpayne@68 16 * @date Jul 5, 2012
jpayne@68 17 *
jpayne@68 18 */
jpayne@68 19 public class KmerCount4 extends KmerCountAbstract {
jpayne@68 20
jpayne@68 21 public static void main(String[] args){
jpayne@68 22
jpayne@68 23 Timer t=new Timer();
jpayne@68 24
jpayne@68 25 String fname1=args[0];
jpayne@68 26 String fname2=(args.length>3 || args[1].contains(".") ? args[1] : null);
jpayne@68 27 int k=14;
jpayne@68 28 int cbits=16;
jpayne@68 29 int gap=0;
jpayne@68 30
jpayne@68 31 for(int i=(fname2==null ? 1 : 2); i<args.length; i++){
jpayne@68 32 final String arg=args[i];
jpayne@68 33 final String[] split=arg.split("=");
jpayne@68 34 String a=split[0].toLowerCase();
jpayne@68 35 String b=split.length>1 ? split[1] : null;
jpayne@68 36
jpayne@68 37 if(a.equals("k") || a.equals("kmer")){
jpayne@68 38 k=Integer.parseInt(b);
jpayne@68 39 }else if(a.startsWith("cbits") || a.startsWith("cellbits")){
jpayne@68 40 cbits=Integer.parseInt(b);
jpayne@68 41 }else if(a.startsWith("gap")){
jpayne@68 42 gap=Integer.parseInt(b);
jpayne@68 43 }else{
jpayne@68 44 throw new RuntimeException("Unknown parameter "+args[i]);
jpayne@68 45 }
jpayne@68 46 }
jpayne@68 47
jpayne@68 48 KCountArray2 count=null;
jpayne@68 49
jpayne@68 50 if(fileIO.FileFormat.hasFastaExtension(fname1)){
jpayne@68 51 assert(!FastaReadInputStream.SPLIT_READS);
jpayne@68 52 FastaReadInputStream.MIN_READ_LEN=k;
jpayne@68 53 }
jpayne@68 54
jpayne@68 55 if(gap==0){
jpayne@68 56 count=count(fname1, fname2, k, cbits, true);
jpayne@68 57 }else{
jpayne@68 58 count=countFastqSplit(fname1, fname2, (k+1)/2, k/2, gap, cbits, true, null);
jpayne@68 59 }
jpayne@68 60
jpayne@68 61
jpayne@68 62 t.stop();
jpayne@68 63 System.out.println("Finished counting; time = "+t);
jpayne@68 64
jpayne@68 65 printStatistics(count);
jpayne@68 66
jpayne@68 67 }
jpayne@68 68
jpayne@68 69 public static void printStatistics(KCountArray2 count){
jpayne@68 70 long[] freq=count.transformToFrequency();
jpayne@68 71
jpayne@68 72 // System.out.println(count+"\n");
jpayne@68 73 // System.out.println(Arrays.toString(freq)+"\n");
jpayne@68 74
jpayne@68 75 long sum=sum(freq);
jpayne@68 76 System.out.println("Kmer fraction:");
jpayne@68 77 int lim1=8, lim2=16;
jpayne@68 78 for(int i=0; i<lim1; i++){
jpayne@68 79 String prefix=i+"";
jpayne@68 80 while(prefix.length()<8){prefix=prefix+" ";}
jpayne@68 81 System.out.println(prefix+"\t"+String.format(Locale.ROOT, "%.3f%% ",(100l*freq[i]/(double)sum))+"\t"+freq[i]);
jpayne@68 82 }
jpayne@68 83 while(lim1<=freq.length){
jpayne@68 84 int x=0;
jpayne@68 85 for(int i=lim1; i<lim2; i++){
jpayne@68 86 x+=freq[i];
jpayne@68 87 }
jpayne@68 88 String prefix=lim1+"-"+(lim2-1);
jpayne@68 89 if(lim2>=freq.length){prefix=lim1+"+";}
jpayne@68 90 while(prefix.length()<8){prefix=prefix+" ";}
jpayne@68 91 System.out.println(prefix+"\t"+String.format(Locale.ROOT, "%.3f%% ",(100l*x/(double)sum))+"\t"+x);
jpayne@68 92 lim1*=2;
jpayne@68 93 lim2=min(lim2*2, freq.length);
jpayne@68 94 }
jpayne@68 95
jpayne@68 96 long sum2=sum-freq[0];
jpayne@68 97 long x=freq[1];
jpayne@68 98 System.out.println();
jpayne@68 99 System.out.println("Keys Counted: \t \t"+keysCounted);
jpayne@68 100 System.out.println("Unique: \t \t"+sum2);
jpayne@68 101 System.out.println("Avg Sites/Key: \t \t"+String.format(Locale.ROOT, "%.3f ",(keysCounted*1d/sum2)));
jpayne@68 102 System.out.println();
jpayne@68 103 System.out.println("Singleton: \t"+String.format(Locale.ROOT, "%.3f%% ",(100l*x/(double)sum2))+"\t"+x);
jpayne@68 104 x=sum2-x;
jpayne@68 105 System.out.println("Useful: \t"+String.format(Locale.ROOT, "%.3f%% ",(100l*x/(double)sum2))+"\t"+x);
jpayne@68 106 }
jpayne@68 107
jpayne@68 108 public static KCountArray2 count(String reads1, String reads2, int k, int cbits, boolean rcomp){
jpayne@68 109 return count(reads1, reads2, k, cbits, rcomp, null);
jpayne@68 110 }
jpayne@68 111
jpayne@68 112 public static KCountArray2 count(String reads1, String reads2, int k, int cbits, boolean rcomp, KCountArray2 count){
jpayne@68 113 assert(k>=1 && k<20);
jpayne@68 114 final int kbits=2*k;
jpayne@68 115 final long mask=(kbits>63 ? -1L : ~((-1L)<<kbits));
jpayne@68 116
jpayne@68 117 if(count==null){
jpayne@68 118 final long cells=1L<<kbits;
jpayne@68 119 if(verbose){System.err.println("k="+k+", kbits="+kbits+", cells="+cells+", mask="+Long.toHexString(mask));}
jpayne@68 120 count=new KCountArray2(cells, cbits);
jpayne@68 121 }
jpayne@68 122
jpayne@68 123 final ConcurrentReadInputStream cris;
jpayne@68 124 {
jpayne@68 125 FileFormat ff1=FileFormat.testInput(reads1, FileFormat.FASTQ, null, true, true);
jpayne@68 126 FileFormat ff2=FileFormat.testInput(reads2, FileFormat.FASTQ, null, true, true);
jpayne@68 127 cris=ConcurrentReadInputStream.getReadInputStream(maxReads, true, ff1, ff2);
jpayne@68 128 cris.start(); //4567
jpayne@68 129 }
jpayne@68 130
jpayne@68 131 assert(cris!=null) : reads1;
jpayne@68 132 System.err.println("Started cris");
jpayne@68 133 boolean paired=cris.paired();
jpayne@68 134 if(verbose){System.err.println("Paired: "+paired);}
jpayne@68 135
jpayne@68 136 ListNum<Read> ln=cris.nextList();
jpayne@68 137 ArrayList<Read> reads=(ln!=null ? ln.list : null);
jpayne@68 138
jpayne@68 139 if(reads!=null && !reads.isEmpty()){
jpayne@68 140 Read r=reads.get(0);
jpayne@68 141 assert(paired==(r.mate!=null));
jpayne@68 142 }
jpayne@68 143
jpayne@68 144 while(ln!=null && reads!=null && reads.size()>0){//ln!=null prevents a compiler potential null access warning
jpayne@68 145 //System.err.println("reads.size()="+reads.size());
jpayne@68 146 for(Read r : reads){
jpayne@68 147 readsProcessed++;
jpayne@68 148
jpayne@68 149 addRead(r, count, k, mask, rcomp);
jpayne@68 150 if(r.mate!=null){
jpayne@68 151 addRead(r.mate, count, k, mask, rcomp);
jpayne@68 152 }
jpayne@68 153
jpayne@68 154 }
jpayne@68 155 //System.err.println("returning list");
jpayne@68 156 cris.returnList(ln);
jpayne@68 157 //System.err.println("fetching list");
jpayne@68 158 ln=cris.nextList();
jpayne@68 159 reads=(ln!=null ? ln.list : null);
jpayne@68 160 }
jpayne@68 161 if(verbose){System.err.println("Finished reading");}
jpayne@68 162 cris.returnList(ln);
jpayne@68 163 if(verbose){System.err.println("Returned list");}
jpayne@68 164 cris.close();
jpayne@68 165 if(verbose){System.err.println("Closed stream");}
jpayne@68 166 if(verbose){System.err.println("Processed "+readsProcessed+" reads.");}
jpayne@68 167
jpayne@68 168
jpayne@68 169 return count;
jpayne@68 170 }
jpayne@68 171
jpayne@68 172 public static KCountArray2 countFastqSplit(String reads1, String reads2, int k1, int k2, int gap, int cbits, boolean rcomp, KCountArray2 count){
jpayne@68 173 assert(k1+k2>=1 && k1+k2<20);
jpayne@68 174 assert(gap>=0);
jpayne@68 175 final int kbits1=2*k1;
jpayne@68 176 final int kbits2=2*k2;
jpayne@68 177 final long mask1=~((-1L)<<(kbits1));
jpayne@68 178 final long mask2=~((-1L)<<(kbits2));
jpayne@68 179
jpayne@68 180 if(count==null){
jpayne@68 181 final long cells=1L<<(kbits1+kbits2);
jpayne@68 182 if(verbose){System.err.println("k1="+k1+", k2="+k2+", kbits1="+kbits1+", kbits2="+kbits2+", cells="+cells+
jpayne@68 183 ", mask1="+Long.toHexString(mask1)+", mask2="+Long.toHexString(mask2));}
jpayne@68 184 count=new KCountArray2(cells, cbits, gap);
jpayne@68 185 }
jpayne@68 186 assert(count.gap==gap);
jpayne@68 187
jpayne@68 188 final ConcurrentReadInputStream cris;
jpayne@68 189 {
jpayne@68 190 FileFormat ff1=FileFormat.testInput(reads1, FileFormat.FASTQ, null, true, true);
jpayne@68 191 FileFormat ff2=FileFormat.testInput(reads2, FileFormat.FASTQ, null, true, true);
jpayne@68 192 cris=ConcurrentReadInputStream.getReadInputStream(maxReads, true, ff1, ff2);
jpayne@68 193 cris.start(); //4567
jpayne@68 194 }
jpayne@68 195
jpayne@68 196 assert(cris!=null) : reads1;
jpayne@68 197 System.err.println("Started cris");
jpayne@68 198 boolean paired=cris.paired();
jpayne@68 199 if(verbose){System.err.println("Paired: "+paired);}
jpayne@68 200
jpayne@68 201 ListNum<Read> ln=cris.nextList();
jpayne@68 202 ArrayList<Read> reads=(ln!=null ? ln.list : null);
jpayne@68 203
jpayne@68 204 if(reads!=null && !reads.isEmpty()){
jpayne@68 205 Read r=reads.get(0);
jpayne@68 206 assert(paired==(r.mate!=null));
jpayne@68 207 }
jpayne@68 208
jpayne@68 209 while(ln!=null && reads!=null && reads.size()>0){//ln!=null prevents a compiler potential null access warning
jpayne@68 210 //System.err.println("reads.size()="+reads.size());
jpayne@68 211 for(Read r : reads){
jpayne@68 212 readsProcessed++;
jpayne@68 213
jpayne@68 214 addReadSplit(r, count, k1, k2, mask1, mask2, gap, rcomp);
jpayne@68 215 if(r.mate!=null){
jpayne@68 216 addReadSplit(r.mate, count, k1, k2, mask1, mask2, gap, rcomp);
jpayne@68 217 }
jpayne@68 218
jpayne@68 219 }
jpayne@68 220 //System.err.println("returning list");
jpayne@68 221 cris.returnList(ln);
jpayne@68 222 //System.err.println("fetching list");
jpayne@68 223 ln=cris.nextList();
jpayne@68 224 reads=(ln!=null ? ln.list : null);
jpayne@68 225 }
jpayne@68 226 if(verbose){System.err.println("Finished reading");}
jpayne@68 227 cris.returnList(ln);
jpayne@68 228 if(verbose){System.err.println("Returned list");}
jpayne@68 229 cris.close();
jpayne@68 230 if(verbose){System.err.println("Closed stream");}
jpayne@68 231 if(verbose){System.err.println("Processed "+readsProcessed+" reads.");}
jpayne@68 232
jpayne@68 233
jpayne@68 234 return count;
jpayne@68 235 }
jpayne@68 236
jpayne@68 237 public static void addRead(final Read r, final KCountArray2 count, final int k, final long mask, boolean rcomp){
jpayne@68 238 int len=0;
jpayne@68 239 long kmer=0;
jpayne@68 240 byte[] bases=r.bases;
jpayne@68 241 byte[] quals=r.quality;
jpayne@68 242 for(int i=0; i<bases.length; i++){
jpayne@68 243 byte b=bases[i];
jpayne@68 244 int x=AminoAcid.baseToNumber[b];
jpayne@68 245 if(x<0 || (quals!=null && quals[i]<minQuality)){
jpayne@68 246 len=0;
jpayne@68 247 kmer=0;
jpayne@68 248 }else{
jpayne@68 249 kmer=((kmer<<2)|x)&mask;
jpayne@68 250 len++;
jpayne@68 251 if(len>=k){
jpayne@68 252 keysCounted++;
jpayne@68 253 // System.out.print("Incrementing "+Long.toHexString(kmer)+": "+count.read(kmer));
jpayne@68 254 count.increment(kmer, 1);
jpayne@68 255 // System.out.println(" -> "+count.read(kmer));
jpayne@68 256 // System.out.print("Incrementing array for "+Long.toHexString(kmer)+": "+array[(int)kmer]);
jpayne@68 257 // array[(int)kmer]++;
jpayne@68 258 // System.out.println(" -> "+array[(int)kmer]+"\n");
jpayne@68 259 // assert(array[(int)kmer]==count.read(kmer) || array[(int)kmer]>3);
jpayne@68 260 }
jpayne@68 261 }
jpayne@68 262 }
jpayne@68 263 if(rcomp){
jpayne@68 264 r.reverseComplement();
jpayne@68 265 addRead(r, count, k, mask, false);
jpayne@68 266 }
jpayne@68 267 }
jpayne@68 268
jpayne@68 269 public static void addReadSplit(final Read r, final KCountArray2 count, final int k1, final int k2, final long mask1, final long mask2, final int gap, boolean rcomp){
jpayne@68 270 int len=0;
jpayne@68 271 int shift=k2*2;
jpayne@68 272 long kmer1=0;
jpayne@68 273 long kmer2=0;
jpayne@68 274 byte[] bases=r.bases;
jpayne@68 275 byte[] quals=r.quality;
jpayne@68 276
jpayne@68 277 assert(kmer1>=kmer2);
jpayne@68 278
jpayne@68 279 // assert(false) : k1+", "+k2+", "+mask1+", "+mask2+", "+gap;
jpayne@68 280
jpayne@68 281 for(int i=0, j=i+k1+gap; j<bases.length; i++, j++){
jpayne@68 282 int x1=AminoAcid.baseToNumber[bases[i]];
jpayne@68 283 int x2=AminoAcid.baseToNumber[bases[j]];
jpayne@68 284 if(x1<0 || x2<0 || (quals!=null && (quals[i]<minQuality || quals[j]<minQuality))){
jpayne@68 285 len=0;
jpayne@68 286 kmer1=0;
jpayne@68 287 kmer2=0;
jpayne@68 288 }else{
jpayne@68 289 kmer1=((kmer1<<2)|x1)&mask1;
jpayne@68 290 kmer2=((kmer2<<2)|x2)&mask2;
jpayne@68 291 len++;
jpayne@68 292 if(len>=k1){
jpayne@68 293 keysCounted++;
jpayne@68 294 // System.out.print("Incrementing "+Long.toHexString(kmer)+": "+count.read(kmer));
jpayne@68 295
jpayne@68 296 long key=(kmer1<<shift)|kmer2;
jpayne@68 297 // System.err.println(Long.toHexString(key));
jpayne@68 298 count.increment(key, 1);
jpayne@68 299 // System.out.println(" -> "+count.read(kmer));
jpayne@68 300 // System.out.print("Incrementing array for "+Long.toHexString(kmer)+": "+array[(int)kmer]);
jpayne@68 301 // array[(int)kmer]++;
jpayne@68 302 // System.out.println(" -> "+array[(int)kmer]+"\n");
jpayne@68 303 // assert(array[(int)kmer]==count.read(kmer) || array[(int)kmer]>3);
jpayne@68 304 }
jpayne@68 305 }
jpayne@68 306 }
jpayne@68 307 if(rcomp){
jpayne@68 308 r.reverseComplement();
jpayne@68 309 addReadSplit(r, count, k1, k2, mask1, mask2, gap, false);
jpayne@68 310 }
jpayne@68 311 }
jpayne@68 312
jpayne@68 313 public static void addReadSplit(final byte[] bases, final KCountArray2 count, final int k1, final int k2, final long mask1, final long mask2, final int gap, boolean rcomp){
jpayne@68 314 int len=0;
jpayne@68 315 int shift=k2*2;
jpayne@68 316 long kmer1=0;
jpayne@68 317 long kmer2=0;
jpayne@68 318 byte[] quals=null;
jpayne@68 319
jpayne@68 320 assert(kmer1>=kmer2);
jpayne@68 321
jpayne@68 322 // assert(false) : k1+", "+k2+", "+mask1+", "+mask2+", "+gap;
jpayne@68 323
jpayne@68 324 for(int i=0, j=i+k1+gap; j<bases.length; i++, j++){
jpayne@68 325 int x1=AminoAcid.baseToNumber[bases[i]];
jpayne@68 326 int x2=AminoAcid.baseToNumber[bases[j]];
jpayne@68 327 if(x1<0 || x2<0 || (quals!=null && (quals[i]<minQuality || quals[j]<minQuality))){
jpayne@68 328 len=0;
jpayne@68 329 kmer1=0;
jpayne@68 330 kmer2=0;
jpayne@68 331 }else{
jpayne@68 332 kmer1=((kmer1<<2)|x1)&mask1;
jpayne@68 333 kmer2=((kmer2<<2)|x2)&mask2;
jpayne@68 334 len++;
jpayne@68 335 if(len>=k1){
jpayne@68 336 keysCounted++;
jpayne@68 337 // System.out.print("Incrementing "+Long.toHexString(kmer)+": "+count.read(kmer));
jpayne@68 338
jpayne@68 339 long key=(kmer1<<shift)|kmer2;
jpayne@68 340 System.out.println(Long.toHexString(kmer1));
jpayne@68 341 System.out.println(Long.toHexString(kmer2));
jpayne@68 342 System.out.println(Long.toHexString(key));
jpayne@68 343 count.increment(key, 1);
jpayne@68 344 // System.out.println(" -> "+count.read(kmer));
jpayne@68 345 // System.out.print("Incrementing array for "+Long.toHexString(kmer)+": "+array[(int)kmer]);
jpayne@68 346 // array[(int)kmer]++;
jpayne@68 347 // System.out.println(" -> "+array[(int)kmer]+"\n");
jpayne@68 348 // assert(array[(int)kmer]==count.read(kmer) || array[(int)kmer]>3);
jpayne@68 349 }
jpayne@68 350 }
jpayne@68 351 }
jpayne@68 352 if(rcomp){
jpayne@68 353 AminoAcid.reverseComplementBasesInPlace(bases);
jpayne@68 354 addReadSplit(bases, count, k1, k2, mask1, mask2, gap, false);
jpayne@68 355 }
jpayne@68 356 }
jpayne@68 357
jpayne@68 358 }