Debian Codegraph @

bwa-apache2/NEWS.md

raw
1 Release 0.7.18 (14 April 2024)
2 ------------------------------
3
4 Notable changes:
5
6 * Support ARM64 (#359)
7 * Output SAM header line (#336)
8 * Added the XB tag to output alignment score and mapping quality
9 * Fixed a compiling error with GCC 10 (#267)
10 * Avoid potential buffer overflow (#232)
11
12 (0.7.18: 14 April 2024, r1243)
13
14
15
16 Release 0.7.17 (23 October 2017)
17 --------------------------------
18
19 This release adds option -q to preserve the mapping quality of split alignment
20 with a lower alignment score than the primary alignment. Option -5
21 automatically applies -q as well.
22
23 (0.7.17: 23 October 2017, r1188)
24
25
26
27 Release 0.7.16 (30 July 2017)
28 -----------------------------
29
30 This release added a couple of minor features and incorporated multiple pull
31 requests, including:
32
33 * Added option -5, which is useful to some Hi-C pipelines.
34
35 * Fixed an error with samtools sorting (#129). Updated download link for
36 GRCh38 (#123). Fixed README MarkDown formatting (#70). Addressed multiple
37 issues via a collected pull request #139 by @jmarshall. Avoid malformatted
38 SAM header when -R is used with TAB (#84). Output mate CIGAR (#138).
39
40 (0.7.16: 30 July 2017, r1180)
41
42
43
44 Release 0.7.15 (31 May 2016)
45 ----------------------------
46
47 Fixed a long existing bug which potentially leads to underestimated insert size
48 upper bound. This bug should have little effect in practice.
49
50 (0.7.15: 31 May 2016, r1140)
51
52
53
54 Release 0.7.14 (4 May 2016)
55 ---------------------------
56
57 In the ALT mapping mode, this release adds the "AH:*" header tag to SQ lines
58 corresponding to alternate haplotypes.
59
60 (0.7.14: 4 May 2016, r1136)
61
62
63
64 Release 0.7.13 (23 Feburary 2016)
65 ---------------------------------
66
67 This release fixes a few minor bugs in the previous version and adds a few
68 minor features. All BWA algorithms should produce identical output to 0.7.12
69 when there are no ALT contigs.
70
71 Detailed changes:
72
73 * Fixed a bug in "bwa-postalt.js". The old version may produce 0.5% of wrong
74 bases for reads mapped to the ALT contigs.
75
76 * Fixed a potential bug in the multithreading mode. It may occur when mapping
77 is much faster than file reading, which should almost never happen in
78 practice.
79
80 * Changed the download URL of GRCh38.
81
82 * Removed the read overlap mode. It is not working well.
83
84 * Added the ropebwt2 algorithm as an alternative to index large genomes.
85 Ropebwt2 is slower than the "bwtsw" algorithm, but it has a permissive
86 license. This allows us to create an Apache2-licensed BWA (in the "Apache2"
87 branch) for commercial users who are concerned with GPL.
88
89 (0.7.13: 23 Feburary 2016, r1126)
90
91
92
93 Release 0.7.12 (28 December 2014)
94 ---------------------------------
95
96 This release fixed a bug in the pair-end mode when ALT contigs are present. It
97 leads to undercalling in regions overlapping ALT contigs.
98
99 (0.7.12: 28 December 2014, r1039)
100
101
102
103 Release 0.7.11 (23 December, 2014)
104 ----------------------------------
105
106 A major change to BWA-MEM is the support of mapping to ALT contigs in addition
107 to the primary assembly. Part of the ALT mapping strategy is implemented in
108 BWA-MEM and the rest in a postprocessing script for now. Due to the extra
109 layer of complexity on generating the reference genome and on the two-step
110 mapping, we start to provide a wrapper script and precompiled binaries since
111 this release. The package may be more convenient to some specific use cases.
112 For general uses, the single BWA binary still works like the old way.
113
114 Another major addition to BWA-MEM is HLA typing, which made possible with the
115 new ALT mapping strategy. Necessary data and programs are included in the
116 binary release. The wrapper script also optionally performs HLA typing when HLA
117 genes are included in the reference genome as additional ALT contigs.
118
119 Other notable changes to BWA-MEM:
120
121 * Added option `-b` to `bwa index`. This option tunes the batch size used in
122 the construction of BWT. It is advised to use large `-b` for huge reference
123 sequences such as the BLAST *nt* database.
124
125 * Optimized for PacBio data. This includes a change to scoring based on a
126 study done by Aaron Quinlan and a heuristic speedup. Further speedup is
127 possible, but needs more careful investigation.
128
129 * Dropped PacBio read-to-read alignment for now. BWA-MEM is good for finding
130 the best hit, but is not very sensitive to suboptimal hits. Option `-x pbread`
131 is still available, but hidden on the command line. This may be removed in
132 future releases.
133
134 * Added a new pre-setting for Oxford Nanopore 2D reads. LAST is still a little
135 more sensitive on older bacterial data, but bwa-mem is as good on more
136 recent data and is times faster for mapping against mammalian genomes.
137
138 * Added LAST-like seeding. This improves the accuracy for longer reads.
139
140 * Added option `-H` to insert arbitrary header lines.
141
142 * Smarter option `-p`. Given an interleaved FASTQ stream, old bwa-mem identifies
143 the 2i-th and (2i+1)-th reads as a read pair. The new verion identifies
144 adjacent reads with the same read name as a read pair. It is possible to mix
145 single-end and paired-end reads in one FASTQ.
146
147 * Improved parallelization. Old bwa-mem waits for I/O. The new version puts
148 I/O on a separate thread. It performs mapping while reading FASTQ and
149 writing SAM. This saves significant wall-clock time when reading from
150 or writing to a slow Unix pipe.
151
152 With the new release, the recommended way to map Illumina reads to GRCh38 is to
153 use the bwakit binary package:
154
155 bwa.kit/run-gen-ref hs38DH
156 bwa.kit/bwa index hs38DH.fa
157 bwa.kit/run-bwamem -t8 -H -o out-prefix hs38DH.fa read1.fq.gz read2.fq.gz | sh
158
159 Please check bwa.kit/README.md for details and command line options.
160
161 (0.7.11: 23 December 2014, r1034)
162
163
164
165 Release 0.7.10 (13 July, 2014)
166 ------------------------------
167
168 Notable changes to BWA-MEM:
169
170 * Fixed a segmentation fault due to an alignment bridging the forward-reverse
171 boundary. This is a bug.
172
173 * Use the PacBio heuristic to map contigs to the reference genome. The old
174 heuristic evaluates the necessity of full extension for each chain. This may
175 not work in long low-complexity regions. The PacBio heuristic performs
176 SSE2-SW around each short seed. It works better. Note that the heuristic is
177 only applied to long query sequences. For Illumina reads, the output is
178 identical to the previous version.
179
180 (0.7.10: 13 July 2014, r789)
181
182
183
184 Release 0.7.9 (19 May, 2014)
185 ----------------------------
186
187 This release brings several major changes to BWA-MEM. Notably, BWA-MEM now
188 formally supports PacBio read-to-reference alignment and experimentally supports
189 PacBio read-to-read alignment. BWA-MEM also runs faster at a minor cost of
190 accuracy. The speedup is more significant when GRCh38 is in use. More
191 specifically:
192
193 * Support PacBio subread-to-reference alignment. Although older BWA-MEM works
194 with PacBio data in principle, the resultant alignments are frequently
195 fragmented. In this release, we fine tuned existing methods and introduced
196 new heuristics to improve PacBio alignment. These changes are not used by
197 default. Users need to add option "-x pacbio" to enable the feature.
198
199 * Support PacBio subread-to-subread alignment (EXPERIMENTAL). This feature is
200 enabled with option "-x pbread". In this mode, the output only gives the
201 overlapping region between a pair of reads without detailed alignment.
202
203 * Output alternative hits in the XA tag if there are not so many of them. This
204 is a BWA-backtrack feature.
205
206 * Support mapping to ALT contigs in GRCh38 (EXPERIMENTAL). We provide a script
207 to postprocess hits in the XA tag to adjust the mapping quality and generate
208 new primary alignments to all overlapping ALT contigs. We would *NOT*
209 recommend this feature for production uses.
210
211 * Improved alignments to many short reference sequences. Older BWA-MEM may
212 generate an alignment bridging two or more adjacent reference sequences.
213 Such alignments are split at a later step as postprocessing. This approach
214 is complex and does not always work. This release forbids these alignments
215 from the very beginning. BWA-MEM should not produce an alignment bridging
216 two or more reference sequences any more.
217
218 * Reduced the maximum seed occurrence from 10000 to 500. Reduced the maximum
219 rounds of Smith-Waterman mate rescue from 100 to 50. Added a heuristic to
220 lower the mapping quality if a read contains seeds with excessive
221 occurrences. These changes make BWA-MEM faster at a minor cost of accuracy
222 in highly repetitive regions.
223
224 * Added an option "-Y" to use soft clipping for supplementary alignments.
225
226 * Bugfix: incomplete alignment extension in corner cases.
227
228 * Bugfix: integer overflow when aligning long query sequences.
229
230 * Bugfix: chain score is not computed correctly (almost no practical effect)
231
232 * General code cleanup
233
234 * Added FAQs to README
235
236 Changes in BWA-backtrack:
237
238 * Bugfix: a segmentation fault when an alignment stands out of the end of the
239 last chromosome.
240
241 (0.7.9: 19 May 2014, r783)
242
243
244
245 Release 0.7.8 (31 March, 2014)
246 ------------------------------
247
248 Changes in BWA-MEM:
249
250 * Bugfix: off-diagonal X-dropoff (option -d) not working as intended.
251 Short-read alignment is not affected.
252
253 * Bugfix: unnecessarily large bandwidth used during global alignment,
254 which reduces the mapping speed by -5% for short reads. Results are not
255 affected.
256
257 * Bugfix: when the matching score is not one, paired-end mapping quality is
258 inaccurate.
259
260 * When the matching score (option -A) is changed, scale all score-related
261 options accordingly unless overridden by users.
262
263 * Allow to specify different gap open (or extension) penalties for deletions
264 and insertions separately.
265
266 * Allow to specify the insert size distribution.
267
268 * Better and more detailed debugging information.
269
270 With the default setting, 0.7.8 and 0.7.7 gave identical output on one million
271 100bp read pairs.
272
273 (0.7.8: 31 March 2014, r455)
274
275
276
277 Release 0.7.7 (25 Feburary, 2014)
278 ---------------------------------
279
280 This release fixes incorrect MD tags in the BWA-MEM output.
281
282 A note about short-read mapping to GRCh38. The new human reference genome
283 GRCh38 contains 60Mbp program generated alpha repeat arrays, some of which are
284 hard masked as they cannot be localized. These highly repetitive arrays make
285 BWA-MEM -50% slower. If you are concerned with the performance of BWA-MEM, you
286 may consider to use option "-c2000 -m50". On simulated data, this setting helps
287 the performance at a very minor cost on accuracy. I may consider to change the
288 default in future releases.
289
290 (0.7.7: 25 Feburary 2014, r441)
291
292
293
294 Release 0.7.6 (31 Januaray, 2014)
295 ---------------------------------
296
297 Changes in BWA-MEM:
298
299 * Changed the way mapping quality is estimated. The new method tends to give
300 the same alignment a higher mapping quality. On paired-end reads, the change
301 is minor as with pairing, the mapping quality is usually high. For short
302 single-end reads, the difference is considerable.
303
304 * Improved load balance when many threads are spawned. However, bwa-mem is
305 still not very thread efficient, probably due to the frequent heap memory
306 allocation. Further improvement is a little difficult and may affect the
307 code stability.
308
309 * Allow to use different clipping penalties for 5'- and 3'-ends. This helps
310 when we do not want to clip one end.
311
312 * Print the @PG line, including the command line options.
313
314 * Improved the band width estimate: a) fixed a bug causing the band
315 width extimated from extension not used in the final global alignment; b)
316 try doubled band width if the global alignment score is smaller.
317 Insufficient band width leads to wrong CIGAR and spurious mismatches/indels.
318
319 * Added a new option -D to fine tune a heuristic on dropping suboptimal hits.
320 Reducing -D increases accuracy but decreases the mapping speed. If unsure,
321 leave it to the default.
322
323 * Bugfix: for a repetitive single-end read, the reported hit is not randomly
324 distributed among equally best hits.
325
326 * Bugfix: missing paired-end hits due to unsorted list of SE hits.
327
328 * Bugfix: incorrect CIGAR caused by a defect in the global alignment.
329
330 * Bugfix: incorrect CIGAR caused by failed SW rescue.
331
332 * Bugfix: alignments largely mapped to the same position are regarded to be
333 distinct from each other, which leads to underestimated mapping quality.
334
335 * Added the MD tag.
336
337 There are no changes to BWA-backtrack in this release. However, it has a few
338 known issues yet to be fixed. If you prefer BWA-track, It is still advised to
339 use bwa-0.6.x.
340
341 While I developed BWA-MEM, I also found a few issues with BWA-SW. It is now
342 possible to improve BWA-SW with the lessons learned from BWA-MEM. However, as
343 BWA-MEM is usually better, I will not improve BWA-SW until I find applications
344 where BWA-SW may excel.
345
346 (0.7.6: 31 January 2014, r432)
347
348
349
350 Release 0.7.5a (30 May, 2013)
351 -----------------------------
352
353 Fixed a bug in BWA-backtrack which leads to off-by-one mapping errors in rare
354 cases.
355
356 (0.7.5a: 30 May 2013, r405)
357
358
359
360 Release 0.7.5 (29 May, 2013)
361 ----------------------------
362
363 Changes in all components:
364
365 * Improved error checking on memory allocation and file I/O. Patches provided
366 by Rob Davies.
367
368 * Updated README.
369
370 * Bugfix: return code is zero upon errors.
371
372 Changes in BWA-MEM:
373
374 * Changed the way a chimeric alignment is reported (conforming to the upcoming
375 SAM spec v1.5). With 0.7.5, if the read has a chimeric alignment, the paired
376 or the top hit uses soft clipping and is marked with neither 0x800 nor 0x100
377 bits. All the other hits part of the chimeric alignment will use hard
378 clipping and be marked with 0x800 if option "-M" is not in use, or marked
379 with 0x100 otherwise.
380
381 * Other hits part of a chimeric alignment are now reported in the SA tag,
382 conforming to the SAM spec v1.5.
383
384 * Better method for resolving an alignment bridging two or more short
385 reference sequences. The current strategy maps the query to the reference
386 sequence that covers the middle point of the alignment. For most
387 applications, this change has no effects.
388
389 Changes in BWA-backtrack:
390
391 * Added a magic number to .sai files. This prevents samse/sampe from reading
392 corrupted .sai (e.g. a .sai file containing LSF log) or incompatible .sai
393 generated by a different version of bwa.
394
395 * Bugfix: alignments in the XA:Z: tag were wrong.
396
397 * Keep track of #ins and #del during backtracking. This simplifies the code
398 and reduces errors in rare corner cases. I should have done this in the
399 early days of bwa.
400
401 In addition, if you use BWA-MEM or the fastmap command of BWA, please cite:
402
403 - Li H. (2013) Aligning sequence reads, clone sequences and assembly contigs
404 with BWA-MEM. arXiv:1303.3997v2 [q-bio.GN].
405
406 Thank you.
407
408 (0.7.5: 29 May 2013, r404)
409
410
411
412 Release 0.7.4 (23 April, 2013)
413 ------------------------------
414
415 This is a bugfix release. Most of bugs are considered to be minor which only
416 occur very rarely.
417
418 * Bugfix: wrong CIGAR when a query sequence bridges three or more target
419 sequences. This only happens when aligning reads to short assembly contigs.
420
421 * Bugfix: leading "D" operator in CIGAR.
422
423 * Extend more seeds for better alignment around tandem repeats. This is also
424 a cause of the leading "D" operator in CIGAR.
425
426 * Bugfix: SSE2-SSW may occasionally find incorrect query starting position
427 around tandem repeat. This will lead to a suboptimal CIGAR in BWA-MEM and
428 a wrong CIGAR in BWA.
429
430 * Bugfix: clipping penalty does not work as is intended when there is a gap
431 towards the end of a read.
432
433 * Fixed an issue caused by a bug in the libc from Mac/Darwin. In Darwin,
434 fread() is unable to read a data block longer than 2GB due to an integer
435 overflow bug in its implementation.
436
437 Since version 0.7.4, BWA-MEM is considered to reach similar stability to
438 BWA-backtrack for short-read mapping.
439
440 (0.7.4: 23 April, r385)
441
442
443
444 Release 0.7.3a (15 March, 2013)
445 -------------------------------
446
447 In 0.7.3, the wrong CIGAR bug was only fixed in one scenario, but not fixed
448 in another corner case.
449
450 (0.7.3a: 15 March 2013, r367)
451
452
453
454 Release 0.7.3 (15 March, 2013)
455 ------------------------------
456
457 Changes to BWA-MEM:
458
459 * Bugfix: pairing score is inaccurate when option -A does not take the default
460 value. This is a very minor issue even if it happens.
461
462 * Bugfix: occasionally wrong CIGAR. This happens when in the alignment there
463 is a 1bp deletion and a 1bp insertion which are close to the end of the
464 reads, and there are no other substitutions or indels. BWA-MEM would not do
465 a gapped alignment due to the bug.
466
467 * New feature: output other non-overlapping alignments in the XP tag such that
468 we can see the entire picture of alignment from one SAM line. XP gives the
469 position, CIGAR, NM and mapQ of each aligned subsequence of the query.
470
471 BWA-MEM has been used to align -300Gbp 100-700bp SE/PE reads. SNP/indel calling
472 has also been evaluated on part of these data. BWA-MEM generally gives better
473 pre-filtered SNP calls than BWA. No significant issues have been observed since
474 0.7.2, though minor improvements or bugs (e.g. the bug fixed in this release)
475 are still possible. If you find potential issues, please send bug reports to
476 <[email protected]> (free registration required).
477
478 In addition, more detailed description of the BWA-MEM algorithm can be found at
479 <https://github.com/lh3/mem-paper>.
480
481 (0.7.3: 15 March 2013, r366)
482
483
484
485 Release 0.7.2 (9 March, 2013)
486 -----------------------------
487
488 Emergent bug fix: 0.7.0 and 0.7.1 give a wrong sign to TLEN. In addition,
489 flagging 'properly paired' also gets improved a little.
490
491 (0.7.2: 9 March 2013, r351)
492
493
494
495 Release 0.7.1 (8 March, 2013)
496 -----------------------------
497
498 Changes to BWA-MEM:
499
500 * Bugfix: rare segmentation fault caused by a partial hit to the end of the
501 last sequence.
502
503 * Bugfix: occasional mis-pairing given an interleaved fastq.
504
505 * Bugfix: wrong mate information when the mate is unmapped. SAM generated by
506 BWA-MEM can now be validated with Picard.
507
508 * Improved the performance and accuracy for ultra-long query sequences.
509 Short-read alignment is not affected.
510
511 Changes to other components:
512
513 * In BWA-backtrack and BWA-SW, replaced the code for global alignment,
514 Smith-Waterman and SW extension. The performance and accuracy of the two
515 algorithms stay the same.
516
517 * Added an experimental subcommand to merge overlapping paired ends. The
518 algorithm is very conservative: it may miss true overlaps but rarely makes
519 mistakes.
520
521 An important note is that like BWA-SW, BWA-MEM may output multiple primary
522 alignments for a read, which may cause problems to some tools. For aligning
523 sequence reads, it is advised to use '-M' to flag extra hits as secondary. This
524 option is not the default because multiple primary alignments are theoretically
525 possible in sequence alignment.
526
527 (0.7.1: 8 March 2013, r347)
528
529
530
531 Beta Release 0.7.0 (28 Feburary, 2013)
532 --------------------------------------
533
534 This release comes with a new alignment algorithm, BWA-MEM, for 70bp-1Mbp query
535 sequences. BWA-MEM essentially seeds alignments with a variant of the fastmap
536 algorithm and extends seeds with banded affine-gap-penalty dynamic programming
537 (i.e. the Smith-Waterman-Gotoh algorithm). For typical Illumina 100bp reads or
538 longer low-divergence query sequences, BWA-MEM is about twice as fast as BWA
539 and BWA-SW and is more accurate. It also supports split alignments like BWA-SW
540 and may optionally output multiple hits like BWA. BWA-MEM does not guarantee
541 to find hits within a certain edit distance, but BWA is not efficient for such
542 task given longer reads anyway, and the edit-distance criterion is arguably
543 not as important in long-read alignment.
544
545 In addition to the algorithmic improvements, BWA-MEM also implements a few
546 handy features in practical aspects:
547
548 1. BWA-MEM automatically switches between local and glocal (global wrt reads;
549 local wrt reference) alignment. It reports the end-to-end glocal alignment
550 if the glocal alignment is not much worse than the optimal local alignment.
551 Glocal alignment reduces reference bias.
552
553 2. BWA-MEM automatically infers pair orientation from a batch of single-end
554 alignments. It allows more than one orientations if there are sufficient
555 supporting reads. This feature has not been tested on reads from Illumina
556 jumping library yet. (EXPERIMENTAL)
557
558 3. BWA-MEM optionally takes one interleaved fastq for paired-end mapping. It
559 is possible to convert a name-sorted BAM to an interleaved fastq on the fly
560 and feed the data stream to BWA-MEM for mapping.
561
562 4. BWA-MEM optionally copies FASTA/Q comments to the final SAM output, which
563 helps to transfer individual read annotations to the output.
564
565 5. BWA-MEM supports more advanced piping. Users can now run:
566 (bwa mem ref.fa '<bzcat r1.fq.bz2' '<bzcat r2.fq.bz2') to map bzip'd read
567 files without replying on bash features.
568
569 6. BWA-MEM provides a few basic APIs for single-end mapping. The 'example.c'
570 program in the source code directory implements a full single-end mapper in
571 50 lines of code.
572
573 The BWA-MEM algorithm is in the beta phase. It is not advised to use BWA-MEM
574 for production use yet. However, when the implementation becomes stable after a
575 few release cycles, existing BWA users are recommended to migrate to BWA-MEM
576 for 76bp or longer Illumina reads and long query sequences. The original BWA
577 short-read algorithm will not deliver satisfactory results for 150bp+ Illumina
578 reads. Change of mappers will be necessary sooner or later.
579
580 (0.7.0 beta: 28 Feburary 2013, r313)
581
582
583
584 Release 0.6.2 (19 June, 2012)
585 -----------------------------
586
587 This is largely a bug-fix release. Notable changes in BWA-short and BWA-SW:
588
589 * Bugfix: BWA-SW may give bad alignments due to incorrect band width.
590
591 * Bugfix: A segmentation fault due to an out-of-boundary error. The fix is a
592 temporary solution. The real cause has not been identified.
593
594 * Attempt to read index from prefix.64.bwt, such that the 32-bit and 64-bit
595 index can coexist.
596
597 * Added options '-I' and '-S' to control BWA-SW pairing.
598
599 (0.6.2: 19 June 2012, r126)
600
601
602
603 Release 0.6.1 (28 November, 2011)
604 ---------------------------------
605
606 Notable changes to BWA-short:
607
608 * Bugfix: duplicated alternative hits in the XA tag.
609
610 * Bugfix: when trimming enabled, bwa-aln trims 1bp less.
611
612 * Disabled the color-space alignment. 0.6.x is not working with SOLiD reads at
613 present.
614
615 Notable changes to BWA-SW:
616
617 * Bugfix: segfault due to excessive ambiguous bases.
618
619 * Bugfix: incorrect mate position in the SE mode.
620
621 * Bugfix: rare segfault in the PE mode
622
623 * When macro _NO_SSE2 is in use, fall back to the standard Smith-Waterman
624 instead of SSE2-SW.
625
626 * Optionally mark split hits with lower alignment scores as secondary.
627
628 Changes to fastmap:
629
630 * Bugfix: infinite loop caused by ambiguous bases.
631
632 * Optionally output the query sequence.
633
634 (0.6.1: 28 November 2011, r104)
635
636
637
638 Release 0.5.10 and 0.6.0 (12 November, 2011)
639 --------------------------------------------
640
641 The 0.6.0 release comes with two major changes. Firstly, the index data
642 structure has been changed to support genomes longer than 4GB. The forward and
643 reverse backward genome is now integrated in one index. This change speeds up
644 BWA-short by about 20% and BWA-SW by 90% with the mapping acccuracy largely
645 unchanged. A tradeoff is BWA requires more memory, but this is the price almost
646 all mappers that index the genome have to pay.
647
648 Secondly, BWA-SW in 0.6.0 now works with paired-end data. It is more accurate
649 for highly unique reads and more robust to long indels and structural
650 variations. However, BWA-short still has edges for reads with many suboptimal
651 hits. It is yet to know which algorithm is the best for variant calling.
652
653 0.5.10 is a bugfix release only and is likely to be the last release in the 0.5
654 branch unless I find critical bugs in future.
655
656 Other notable changes:
657
658 * Added the 'fastmap' command that finds super-maximal exact matches. It does
659 not give the final alignment, but runs much faster. It can be a building
660 block for other alignment algorithms. [0.6.0 only]
661
662 * Output the timing information before BWA exits. This also tells users that
663 the task has been finished instead of being killed or aborted. [0.6.0 only]
664
665 * Sped up multi-threading when using many (>20) CPU cores.
666
667 * Check I/O error.
668
669 * Increased the maximum barcode length to 63bp.
670
671 * Automatically choose the indexing algorithm.
672
673 * Bugfix: very rare segfault due to an uninitialized variable. The bug also
674 affects the placement of suboptimal alignments. The effect is very minor.
675
676 This release involves quite a lot of tricky changes. Although it has been
677 tested on a few data sets, subtle bugs may be still hidden. It is *NOT*
678 recommended to use this release in a production pipeline. In future, however,
679 BWA-SW may be better when reads continue to go longer. I would encourage users
680 to try the 0.6 release. I would also like to hear the users' experience. Thank
681 you.
682
683 (0.6.0: 12 November 2011, r85)
684
685
686
687 Beta Release 0.5.9 (24 January, 2011)
688 -------------------------------------
689
690 Notable changes:
691
692 * Feature: barcode support via the '-B' option.
693
694 * Feature: Illumina 1.3+ read format support via the '-I' option.
695
696 * Bugfix: RG tags are not attached to unmapped reads.
697
698 * Bugfix: very rare bwasw mismappings
699
700 * Recommend options for PacBio reads in bwasw help message.
701
702
703 Also, since January 13, the BWA master repository has been moved to github:
704
705 https://github.com/lh3/bwa
706
707 The revision number has been reset. All recent changes will be first
708 committed to this repository.
709
710 (0.5.9: 24 January 2011, r16)
711
712
713
714 Beta Release Candidate 0.5.9rc1 (10 December, 2010)
715 ---------------------------------------------------
716
717 Notable changes in bwasw:
718
719 * Output unmapped reads.
720
721 * For a repetitive read, choose a random hit instead of a fixed
722 one. This is not well tested.
723
724 Notable changes in bwa-short:
725
726 * Fixed a bug in the SW scoring system, which may lead to unexpected
727 gaps towards the end of a read.
728
729 * Fixed a bug which invalidates the randomness of repetitive reads.
730
731 * Fixed a rare memory leak.
732
733 * Allowed to specify the read group at the command line.
734
735 * Take name-grouped BAM files as input.
736
737 Changes to this release are usually safe in that they do not interfere
738 with the key functionality. However, the release has only been tested on
739 small samples instead of on large-scale real data. If anything weird
740 happens, please report the bugs to the bio-bwa-help mailing list.
741
742 (0.5.9rc1: 10 December 2010, r1561)
743
744
745
746 Beta Release 0.5.8 (8 June, 2010)
747 ---------------------------------
748
749 Notable changes in bwasw:
750
751 * Fixed an issue of missing alignments. This should happen rarely and
752 only when the contig/read alignment is multi-part. Very rarely, bwasw
753 may still miss a segment in a multi-part alignment. This is difficult
754 to fix, although possible.
755
756 Notable changes in bwa-short:
757
758 * Discard the SW alignment when the best single-end alignment is much
759 better. Such a SW alignment may caused by structural variations and
760 forcing it to be aligned leads to false alignment. This fix has not
761 been tested thoroughly. It would be great to receive more users
762 feedbacks on this issue.
763
764 * Fixed a typo/bug in sampe which leads to unnecessarily large memory
765 usage in some cases.
766
767 * Further reduced the chance of reporting 'weird pairing'.
768
769 (0.5.8: 8 June 2010, r1442)
770
771
772
773 Beta Release 0.5.7 (1 March, 2010)
774 ----------------------------------
775
776 This release only has an effect on paired-end data with fat insert-size
777 distribution. Users are still recommended to update as the new release
778 improves the robustness to poor data.
779
780 * The fix for 'weird pairing' was not working in version 0.5.6, pointed
781 out by Carol Scott. It should work now.
782
783 * Optionally output to a normal file rather than to stdout (by Tim
784 Fennel).
785
786 (0.5.7: 1 March 2010, r1310)
787
788
789
790 Beta Release 0.5.6 (10 Feburary, 2010)
791 --------------------------------------
792
793 Notable changes in bwa-short:
794
795 * Report multiple hits in the SAM format at a new tag XA encoded as:
796 (chr,pos,CIGAR,NM;)*. By default, if a paired or single-end read has
797 4 or fewer hits, they will all be reported; if a read in a anomalous
798 pair has 11 or fewer hits, all of them will be reported.
799
800 * Perform Smith-Waterman alignment also for anomalous read pairs when
801 both ends have quality higher than 17. This reduces false positives
802 for some SV discovery algorithms.
803
804 * Do not report "weird pairing" when the insert size distribution is
805 too fat or has a mean close to zero.
806
807 * If a read is bridging two adjacent chromsomes, flag it as unmapped.
808
809 * Fixed a small but long existing memory leak in paired-end mapping.
810
811 * Multiple bug fixes in SOLiD mapping: a) quality "-1" can be correctly
812 parsed by solid2fastq.pl; b) truncated quality string is resolved; c)
813 SOLiD read mapped to the reverse strand is complemented.
814
815 * Bwa now calculates skewness and kurtosis of the insert size
816 distribution.
817
818 * Deploy a Bayesian method to estimate the maximum distance for a read
819 pair considered to be paired properly. The method is proposed by
820 Gerton Lunter, but bwa only implements a simplified version.
821
822 * Export more functions for Java bindings, by Matt Hanna (See:
823 http://www.broadinstitute.org/gsa/wiki/index.php/Sting_BWA/C_bindings)
824
825 * Abstract bwa CIGAR for further extension, by Rodrigo Goya.
826
827 (0.5.6: 10 Feburary 2010, r1303)
828
829
830
831 Beta Release 0.5.5 (10 November, 2009)
832 --------------------------------------
833
834 This is a bug fix release:
835
836 * Fixed a serious bug/typo in aln which does not occur given short
837 reads, but will lead to segfault for >500bp reads. Of course, the aln
838 command is not recommended for reads longer than 200bp, but this is a
839 bug anyway.
840
841 * Fixed a minor bug/typo which leads to incorrect single-end mapping
842 quality when one end is moved to meet the mate-pair requirement.
843
844 * Fixed a bug in samse for mapping in the color space. This bug is
845 caused by quality filtration added since 0.5.1.
846
847 (0.5.5: 10 November 2009, r1273)
848
849
850
851 Beta Release 0.5.4 (9 October, 2009)
852 ------------------------------------
853
854 Since this version, the default seed length used in the "aln" command is
855 changed to 32.
856
857 Notable changes in bwa-short:
858
859 * Added a new tag "XC:i" which gives the length of clipped reads.
860
861 * In sampe, skip alignments in case of a bug in the Smith-Waterman
862 alignment module.
863
864 * In sampe, fixed a bug in pairing when the read sequence is identical
865 to its reverse complement.
866
867 * In sampe, optionally preload the entire FM-index into memory to
868 reduce disk operations.
869
870 Notable changes in dBWT-SW/BWA-SW:
871
872 * Changed name dBWT-SW to BWA-SW.
873
874 * Optionally use "hard clipping" in the SAM output.
875
876 (0.5.4: 9 October 2009, r1245)
877
878
879
880 Beta Release 0.5.3 (15 September, 2009)
881 ---------------------------------------
882
883 Fixed a critical bug in bwa-short: reads mapped to the reverse strand
884 are not complemented.
885
886 (0.5.3: 15 September 2009, r1225)
887
888
889
890 Beta Release 0.5.2 (13 September, 2009)
891 ---------------------------------------
892
893 Notable changes in bwa-short:
894
895 * Optionally trim reads before alignment. See the manual page on 'aln
896 -q' for detailed description.
897
898 * Fixed a bug in calculating the NM tag for a gapped alignment.
899
900 * Fixed a bug given a mixture of reads with some longer than the seed
901 length and some shorter.
902
903 * Print SAM header.
904
905 Notable changes in dBWT-SW:
906
907 * Changed the default value of -T to 30. As a result, the accuracy is a
908 little higher for short reads at the cost of speed.
909
910 (0.5.2: 13 September 2009, r1223)
911
912
913
914 Beta Release 0.5.1 (2 September, 2009)
915 --------------------------------------
916
917 Notable changes in the short read alignment component:
918
919 * Fixed a bug in samse: do not write mate coordinates.
920
921 Notable changes in dBWT-SW:
922
923 * Randomly choose one alignment if the read is a repetitive.
924
925 * Fixed a flaw when a read is mapped across two adjacent reference
926 sequences. However, wrong alignment reports may still occur rarely in
927 this case.
928
929 * Changed the default band width to 50. The speed is slower due to this
930 change.
931
932 * Improved the mapping quality a little given long query sequences.
933
934 (0.5.1: 2 September 2009, r1209)
935
936
937
938 Beta Release 0.5.0 (20 August, 2009)
939 ------------------------------------
940
941 This release implements a novel algorithm, dBWT-SW, specifically
942 designed for long reads. It is 10-50 times faster than SSAHA2, depending
943 on the characteristics of the input data, and achieves comparable
944 alignment accuracy while allowing chimera detection. In comparison to
945 BLAT, dBWT-SW is several times faster and much more accurate especially
946 when the error rate is high. Please read the manual page for more
947 information.
948
949 The dBWT-SW algorithm is kind of developed for future sequencing
950 technologies which produce much longer reads with a little higher error
951 rate. It is still at its early development stage. Some features are
952 missing and it may be buggy although I have evaluated on several
953 simulated and real data sets. But following the "release early"
954 paradigm, I would like the users to try it first.
955
956 Other notable changes in BWA are:
957
958 * Fixed a rare bug in the Smith-Waterman alignment module.
959
960 * Fixed a rare bug about the wrong alignment coordinate when a read is
961 poorly aligned.
962
963 * Fixed a bug in generating the "mate-unmap" SAM tag when both ends in
964 a pair are unmapped.
965
966 (0.5.0: 20 August 2009, r1200)
967
968
969
970 Beta Release 0.4.9 (19 May, 2009)
971 ---------------------------------
972
973 Interestingly, the integer overflow bug claimed to be fixed in 0.4.7 has
974 not in fact. Now I have fixed the bug. Sorry for this and thank Quan
975 Long for pointing out the bug (again).
976
977 (0.4.9: 19 May 2009, r1075)
978
979
980
981 Beta Release 0.4.8 (18 May, 2009)
982 ---------------------------------
983
984 One change to "aln -R". Now by default, if there are no more than '-R'
985 equally best hits, bwa will search for suboptimal hits. This change
986 affects the ability in finding SNPs in segmental duplications.
987
988 I have not tested this option thoroughly, but this simple change is less
989 likely to cause new bugs. Hope I am right.
990
991 (0.4.8: 18 May 2009, r1073)
992
993
994
995 Beta Release 0.4.7 (12 May, 2009)
996 ---------------------------------
997
998 Notable changes:
999
1000 * Output SM (single-end mapping quality) and AM (smaller mapping
1001 quality among the two ends) tag from sam output.
1002
1003 * Improved the functionality of stdsw.
1004
1005 * Made the XN tag more accurate.
1006
1007 * Fixed a very rare segfault caused by integer overflow.
1008
1009 * Improve the insert size estimation.
1010
1011 * Fixed compiling errors for some Linux systems.
1012
1013 (0.4.7: 12 May 2009, r1066)
1014
1015
1016
1017 Beta Release 0.4.6 (9 March, 2009)
1018 ----------------------------------
1019
1020 This release improves the SOLiD support. First, a script for converting
1021 SOLiD raw data is provided. This script is adapted from solid2fastq.pl
1022 in the MAQ package. Second, a nucleotide reference file can be directly
1023 used with 'bwa index'. Third, SOLiD paired-end support is
1024 completed. Fourth, color-space reads will be converted to nucleotides
1025 when SAM output is generated. Color errors are corrected in this
1026 process. Please note that like MAQ, BWA cannot make use of the primer
1027 base and the first color.
1028
1029 In addition, the calculation of mapping quality is also improved a
1030 little bit, although end-users may barely observe the difference.
1031
1032 (0.4.6: 9 March 2009, r915)
1033
1034
1035
1036 Beta Release 0.4.5 (18 Feburary, 2009)
1037 --------------------------------------
1038
1039 Not much happened, but I think it would be good to let the users use the
1040 latest version.
1041
1042 Notable changes (Thank Bob Handsaker for catching the two bugs):
1043
1044 * Improved bounary check. Previous version may still give incorrect
1045 alignment coordinates in rare cases.
1046
1047 * Fixed a bug in SW alignment when no residue matches. This only
1048 affects the 'sampe' command.
1049
1050 * Robustly estimate insert size without setting the maximum on the
1051 command line. Since this release 'sampe -a' only has an effect if
1052 there are not enough good pairs to infer the insert size
1053 distribution.
1054
1055 * Reduced false PE alignments a little bit by using the inferred insert
1056 size distribution. This fix may be more important for long insert
1057 size libraries.
1058
1059 (0.4.5: 18 Feburary 2009, r829)
1060
1061
1062
1063 Beta Release 0.4.4 (15 Feburary, 2009)
1064 --------------------------------------
1065
1066 This is mainly a bug fix release. Notable changes are:
1067
1068 * Imposed boundary check for extracting subsequence from the
1069 genome. Previously this causes memory problem in rare cases.
1070
1071 * Fixed a bug in failing to find whether an alignment overlapping with
1072 N on the genome.
1073
1074 * Changed MD tag to meet the latest SAM specification.
1075
1076 (0.4.4: 15 Feburary 2009, r815)
1077
1078
1079
1080 Beta Release 0.4.3 (22 January, 2009)
1081 ------------------------------------
1082
1083 Notable changes:
1084
1085 * Treat an ambiguous base N as a mismatch. Previous versions will not
1086 map reads containing any N.
1087
1088 * Automatically choose the maximum allowed number of differences. This
1089 is important when reads of different lengths are mixed together.
1090
1091 * Print mate coordinate if only one end is unmapped.
1092
1093 * Generate MD tag. This tag encodes the mismatching positions and the
1094 reference bases at these positions. Deletions from the reference will
1095 also be printed.
1096
1097 * Optionally dump multiple hits from samse, in another concise format
1098 rather than SAM.
1099
1100 * Optionally disable iterative search. This is VERY SLOOOOW, though.
1101
1102 * Fixed a bug in generate SAM.
1103
1104 (0.4.3: 22 January 2009, r787)
1105
1106
1107
1108 Beta Release 0.4.2 (9 January, 2009)
1109 ------------------------------------
1110
1111 Aaron Quinlan found a bug in the indexer: the bwa indexer segfaults if
1112 there are no comment texts in the FASTA header. This is a critical
1113 bug. Nothing else was changed.
1114
1115 (0.4.2: 9 January 2009, r769)
1116
1117
1118
1119 Beta Release 0.4.1 (7 January, 2009)
1120 ------------------------------------
1121
1122 I am sorry for the quick updates these days. I like to set a milestone
1123 for BWA and this release seems to be. For paired end reads, BWA also
1124 does Smith-Waterman alignment for an unmapped read whose mate can be
1125 mapped confidently. With this strategy BWA achieves similar accuracy to
1126 maq. Benchmark is also updated accordingly.
1127
1128 (0.4.1: 7 January 2009, r760)
1129
1130
1131
1132 Beta Release 0.4.0 (6 January, 2009)
1133 ------------------------------------
1134
1135 In comparison to the release two days ago, this release is mainly tuned
1136 for performance with some tricks I learnt from Bowtie. However, as the
1137 indexing format has also been changed, I have to increase the version
1138 number to 0.4.0 to emphasize that *DATABASE MUST BE RE-INDEXED* with
1139 'bwa index'.
1140
1141 * Improved the speed by about 20%.
1142
1143 * Added multi-threading to 'bwa aln'.
1144
1145 (0.4.0: 6 January 2009, r756)
1146
1147
1148
1149 Beta Release 0.3.0 (4 January, 2009)
1150 ------------------------------------
1151
1152 * Added paired-end support by separating SA calculation and alignment
1153 output.
1154
1155 * Added SAM output.
1156
1157 * Added evaluation to the documentation.
1158
1159 (0.3.0: 4 January 2009, r741)
1160
1161
1162
1163 Beta Release 0.2.0 (15 Augusst, 2008)
1164 -------------------------------------
1165
1166 * Take the subsequence at the 5'-end as seed. Seeding strategy greatly
1167 improves the speed for long reads, at the cost of missing a few true
1168 hits that contain many differences in the seed. Seeding also increase
1169 the memory by 800MB.
1170
1171 * Fixed a bug which may miss some gapped alignments. Fixing the bug
1172 also slows the speed a little.
1173
1174 (0.2.0: 15 August 2008, r428)
1175
1176
1177
1178 Beta Release 0.1.6 (08 Augusst, 2008)
1179 -------------------------------------
1180
1181 * Give accurate CIGAR string.
1182
1183 * Add a simple interface to SW/NW alignment
1184
1185 (0.1.6: 08 August 2008, r414)
1186
1187
1188
1189 Beta Release 0.1.5 (27 July, 2008)
1190 ----------------------------------
1191
1192 * Improve the speed. This version is expected to give the same results.
1193
1194 (0.1.5: 27 July 2008, r400)
1195
1196
1197
1198 Beta Release 0.1.4 (22 July, 2008)
1199 ----------------------------------
1200
1201 * Fixed a bug which may cause missing gapped alignments.
1202
1203 * More clearly define what alignments can be found by BWA (See
1204 manual). Now BWA runs a little slower because it will visit more
1205 potential gapped alignments.
1206
1207 * A bit code clean up.
1208
1209 (0.1.4: 22 July 2008, r387)
1210
1211
1212
1213 Beta Release 0.1.3 (21 July, 2008)
1214 ----------------------------------
1215
1216 Improve the speed with some tricks on retrieving occurrences. The results
1217 should be exactly the same as that of 0.1.2.
1218
1219 (0.1.3: 21 July 2008, r382)
1220
1221
1222
1223 Beta Release 0.1.2 (17 July, 2008)
1224 ----------------------------------
1225
1226 Support gapped alignment. Codes for ungapped alignment has been removed.
1227
1228 (0.1.2: 17 July 2008, r371)
1229
1230
1231
1232 Beta Release 0.1.1 (03 June, 2008)
1233 -----------------------------------
1234
1235 This is the first release of BWA, Burrows-Wheeler Alignment tool. Please
1236 read man page for more information about this software.
1237
1238 (0.1.1: 03 June 2008, r349)