Debian Codegraph @

NEWS

raw
1 Bowtie 2 NEWS
2 =============
3
4 Bowtie 2 is available for download from the project website,
5 http://bowtie-bio.sf.net/bowtie2 and on Github,
6 https://github.com/BenLangmead/bowtie2/releases. 2.0.0-beta1 is
7 the first version released to the public and 2.5.4 is the latest
8 version. Bowtie 2 is licensed under the GPLv3 license. See `LICENSE'
9 file for details.
10
11 Reporting Issues
12 ================
13
14 Please report any issues to the Bowtie 2 Github page or using the Sourceforge bug tracker:
15
16 * https://github.com/BenLangmead/bowtie2/issues
17 * https://sourceforge.net/tracker/?group_id=236897&atid=1101606
18
19 Version Release History
20 =======================
21 ## Version 2.5.5 - Feb 16, 2026 ##
22
23 ### bowtie2 ###
24 * Added `-d` command line option that makes `-a`/`-k`
25 deterministic in the way alignments are reported. This
26 deterministic mode runs much significantly faster than `-a` and
27 consumes less memory. It not yet strictly better than `-a` when
28 considering alignments found. (PR #520, 521)
29 * Add support for selecting AVX version at runtime on x86_64/amd64
30 CPUs. (PR #515)
31
32
33 ## Version 2.5.4 - May 16, 2024 ##
34
35 ### bowtie2 ###
36 * Added `--sam-opt-config` command line option for toggling SAM Opt flags.
37 See MANUAL for details.
38 * Fixed an issue causing `bowtie2`'s memory usage to increase over time
39 when aligning BAM files.
40 * Changed bowtie2 to continue flushing output in the event of a partial write.
41 * Changed the behavior of `bowtie2-build` to throw an exception if it is unable
42 to write the BWT (.1.bt2, .1.rev.bt2). In prior versions `bowtie2-build`, would silently
43 ignore the error which has led some to report the absence of the BWT files in
44 a "completed" index build.
45 * Reverted the changes made in v2.5.0 that sometimes caused unique concordant alignments to be overcounted.
46
47 ## Version 2.5.3 - Jan 16, 2024 ##
48
49 * Fixed an issue causing `bowtie2`'s memory usage to increase over time.
50 * Fixed an issue causing `bowtie2` to crash when processing empty FASTQ records.
51 * Fixed an issue that would cause `bowtie2` to erroneously treat `-f` and `-F` as mutually exclusive.
52 * Add optional support for `libsais` to `bowtie2-build` for faster index building.
53 N.B. -- small indexes are capped at 2GB when using `libsais` as opposed to 4GB when using the
54 default blockwise algorithm. See MANUAL for more details.
55
56 ## Version 2.5.2 - Oct 13, 2023 ##
57
58 ### bowtie2 ###
59 * Fixed an issue preventing `bowtie2` from utilizing the specified number of CPU cores
60 * Fixed an issue causing `bowtie2` to segfault when reading compressed inputs on Windows
61 * Fixed an issue causing `bowtie2` to segfault while parsing interleaved reads
62 * Overhauled the FASTQ parser with better tolerance for empty lines
63
64 ## Version 2.5.1 - Feb 17, 2023 ##
65
66 ### bowtie2 ###
67 * Fixed an issue affecting `bowtie2` alignment accuracy.
68 * Updated the versions of the NCBI `NGS` and `VDB` libraries used by `bowtie2`.
69 * Fixed a segmentation fault that would occur while aligning SRA data.
70 * Fixed an issue preventing `--un-mates` from properly saving unaligned mates.
71
72 ## Version 2.5.0 - Oct 31, 2022 ##
73
74 ### bowtie2 ###
75 * Overall improvements in the use of prefetch instructions. (contribution by Igor Sfiligoi)
76 * Made input/output fully asynchronous by using a dedicated thread. (contribution by Igor Sfiligoi)
77 * Added support for AVX2 256-bit instructions with can be enabled by
78 setting the `SSE_AXV2` environment variable at compile time. (contribution by Igor Sfiligoi)
79 * Fixed an issue causing `bowtie2` to crash when processing ZSTD
80 files with high compression ratios.
81 * Changed the way that unique alignments are counted in summary
82 message to better match up with filters on SAM output
83
84 ## Version 2.4.5 - Jan 15, 2022 ##
85
86 ### bowtie2 ###
87
88 * Fixed issues with `bowtie2` BAM parser that would cause `bowtie2` to
89 crash when processing input that was encoded with tools other than
90 samtools e.g. Picard.
91 * Fixed an issue causing `bowtie2` to drop certain optional fields when
92 when aligning BAM reads with the `--preserve-tags` option.
93 * Fixed an issue causing `bowtie2` to produce mangled SAM output when
94 specifying `--sam-append-comment` together with the `--passthrough`
95 option.
96 * Appended `GO:query` to SAM `@HD` entry to indicate that reads are
97 grouped by query name, bump SAM version to 1.5 to indicate support
98 for this change.
99
100 ### bowtie2-build ###
101
102 * Implemented thread pool to address performance regressions
103 introduced during the switch to C++11 threads.
104 * Fixed an issue causing masked-sequence metadata to be omitted
105 from index. This issue would subsequently result in sequence
106 data, `@SQ`, being left out from alignment SAM header.
107 * Included `.tmp` extension to index files currenlty being built.
108 The extension is removed only upon successful build. This change
109 seeks to address the assumption that `bowtie2-build` ran successfully
110 without building the reverse indexes.
111 * Fixed an issue causing `bowtie2-build` to sometimes incorrectly
112 calculate input size. This issue would result in the wrong index type
113 being chosen and only happened with GZip compressed files.
114
115 ### bowtie2-inspect ###
116
117 * Added a new `-o/--output` option to save the output of
118 `bowtie2-inspect` to a file instead of being dumped to standard
119 output.
120
121
122 ## Version 2.4.4 - May 23, 2021 ##
123
124 * Fixed an issue that would sometimes cause deadlocks in `bowtie2`
125 when running multithreaded
126
127 ## Version 2.4.3 - May 13, 2021 ##
128
129 * Replaced TBB concurrency with C++ threads
130 * Added native support for processing Zstd-compressed read files
131 to bowtie2
132 * Added native support for processing Zstd-compressed
133 reference-genome files to bowtie2-build
134 * Fixed an issue causing `bowtie2` to report incorrect alignments
135 on big-endian machines
136 * Fixed an issue causing `bowtie2` to incorrectly process BAM
137 files on big-endian machines
138 * Fixed an issue causing Bowtie2 to set an incorrect MAPQ when AS and
139 XS are the maximum for read length
140 * Add support for building on Apple M1 processors
141
142 ## Version 2.4.2 - Oct 5, 2020 ##
143
144 * Fixed an issue that would cause the `bowtie2` wrapper script to throw
145 an error when using wrapper-specific arguments.
146 * Added new `--sam-append-comment` flag that appends comment from FASTA/Q
147 read to corresponding SAM record.
148 * Fixed an issue that would cause qupto, `-u`, to overflow when there
149 are >= 2^32 query sequences (PR #312).
150 * Fixed an issue that would cause bowtie2-build script to incorrectly
151 process reference files.
152
153 ## Version 2.4.1 - Feb 28, 2020 ##
154
155 * Fixed an issue that would cause the `bowtie2` wrapper script to incorrectly
156 process certain arguments.
157
158 ## Version 2.4.0 - Feb 25, 2020 ##
159
160 * Fixed an issue in `-b <bam>` input mode where one might prematurely close the
161 read file pointer causing "Bad file descriptor" in other threads
162 * Fixed an issue that could cause `bowtie2` to crash in `--no-1mm-upfront` mode
163 * Modified `bowtie2-build` to better handle of flags and positional parameters
164 * Migrated all `python` scripts to `python3`
165 * Added support for wildcards in input files to `bowtie2`, e.g. `bowtie2 -x
166 index -q *.fq` as opposed to `bowtie2 -x index -q 1.fq,2.fq,3.fq...`
167 * Fixed an issue causing <code>bowtie2</code> to incorrectly process read names
168 with slash mates plus extra characters (see
169 [#265](https://github.com/BenLangmead/bowtie2/issues/265))
170 * Clarified support for overriding presets with more specific options e.g
171 `bowtie2 -x index --local --very-fast-local --L22 -q reads.fq` will set the
172 seed length to 22, overriding the 25 set by `--very-fast-local`
173 * Modified SAM output for `-k`/`-a` so that supplementary alignments get
174 assigned a MAPQ of 255
175 * Fixed an issue that would sometimes cause `bowtie2-build` to not generate
176 reverse index files
177 * Added preliminary support for ppc64le architectures with the help of SIMDE
178 project (see [#271](https://github.com/BenLangmead/bowtie2/issues/271))
179 * Fixed an issue causing `bowtie2` to incorrectly calculate the MAPQ when `--mp`
180 was specified in combination with `--ignore-quals`
181
182 ## Version 2.3.5.1 - Apr 16, 2019 ##
183
184 * Added official support for BAM input files
185 * Added official support for CMake build system
186 * Added changes to Makefile for creating Reproducible builds (via [210](https://github.com/BenLangmead/bowtie2/pull/210))
187 * Fix an issue whereby building on aarch64 would require patching sed commands (via [#243](https://github.com/BenLangmead/bowtie2/pull/243))
188 * Fix an issue whereby `bowtie2` would incorrectly throw an error while processing `--interleaved` input
189
190 ## Version 2.3.5 - Mar 16, 2019 ##
191
192 * Added support for obtaining input reads directly from the Sequence Read Archive, via NCBI's [NGS language bindings](https://github.com/ncbi/ngs). This is activated via the [`--sra-acc`](manual.shtml#bowtie2-options-sra-acc) option. This implementation is based on Daehwan Kim's in [HISAT2](https://ccb.jhu.edu/software/hisat2). Supports both unpaired and paired-end inputs.
193 * Bowtie 2 now compiles on ARM architectures (via [#216](https://github.com/BenLangmead/bowtie2/pull/216))
194 * `--interleaved` can now be combined with FASTA inputs (worked only with FASTQ before)
195 * Fixed issue whereby large indexes were not successfully found in the `$BOWTIE2_INDEXES` directory
196 * Fixed input from FIFOs (e.g. via process substitution) to distinguish gzip-compressed versus uncompressed input
197 * Fixed issue whereby arguments containing `bz2` `lz4` were misinterpretted as files
198 * Fixed several compiler warnings
199 * Fixed issue whereby both ends of a paired-end read could have negative TLEN if they exactly coincided
200 * Fixed issue whereby `bowtie2-build` would hang on end-of-file (via [#228](https://github.com/BenLangmead/bowtie2/pull/228))
201 * Fixed issue whereby wrapper script would sometimes create zombie processes (via [#51](https://github.com/BenLangmead/bowtie2/pull/51))
202 * Fixed issue whereby `bowtie2-build` and `bowtie2-inspect` wrappers would fail on some versions of Python/PyPy
203 * Replaced old, unhelpful `README.md` in the project with a version that includes badges, links and some highlights from the manual
204 * Note: BAM input support and CMake build support both remain experimental, but we expect to finalize them in the next release
205
206 ## Version 2.3.4.3 - Sep 17, 2018 ##
207
208 * Fixed an issue causing `bowtie2-build` and `bowtie2-inspect`
209 to output incomplete help text.
210 * Fixed an issue causing `bowtie2-inspect` to crash.
211 * Fixed an issue preventing `bowtie2` from processing paired and/or
212 unpaired FASTQ reads together with interleaved FASTQ reads.
213
214 ## Version 2.3.4.2 - Aug 7, 2018 ##
215
216 * Fixed issue causing `bowtie2` to fail in `--fast-local` mode.
217 * Fixed issue causing `--soft-clipped-unmapped-tlen` to be a positional argument.
218 * New option `--trim-to N` causes `bowtie2` to trim reads longer
219 than `N` bases to exactly `N` bases. Can trim from either 3'
220 or 5' end, e.g. `--trim-to 5:30` trims reads to 30 bases,
221 truncating at the 5' end.
222 * Updated "Building from source" manual section with additional
223 instructions on installing TBB.
224 * Several other updates to manual, including new mentions of
225 [Bioconda](http://bioconda.github.io) and
226 [Biocontainers](https://biocontainers.pro).
227 * Fixed an issue preventing `bowtie2` from processing more than
228 one pattern source when running single threaded.
229 * Fixed an issue causing `bowtie2` and `bowtie2-inspect` to crash if
230 the index contains a gap-only segment.
231 * Added experimental BAM input mode `-b`. Works only with unpaired
232 input reads and BAM files that are sorted by read name (`samtools
233 sort -n`). BAM input mode also supports the following options:
234 o `--preserve-sam-tags`: Preserve any optional fields present in BAM record
235 o `--align-paired-reads`: Paired-end mode for BAM files
236 * Added experimental cmake support
237
238 Thread-scaling paper appears - July 19, 2018
239
240 * Our latest work on Bowtie's core thread scaling capabilities
241 [just appeared Open Access in the journal Bioinformatics](href="https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/bty648/5055585)
242
243 ## Version 2.3.4.1 - Feb 3, 2018 ##
244
245 * Fixed an issue with `--reorder` that caused bowtie2 to crash
246 while reordering SAM output
247
248 ## Version 2.3.4 - Dec 29, 2017 ##
249
250 * Fixed major issue causing corrupt SAM output when using many
251 threads (-p/--threads) on certain systems.
252 * Fixed an issue whereby bowtie2 processes could overwrite each
253 others' named pipes on HPC systems.
254 * Fixed an issue causing bowtie2-build and bowtie2-inspect to
255 return prematurely on Windows.
256 * Fixed issues raised by compiler "sanitizers" that could
257 potentially have caused memory corruption or undefined behavior.
258 * Added the "continuous FASTA" input format (`-F`) for aligning
259 all the k-mers in the sequences of a FASTA file. Useful for
260 determining mapability of regions of the genome, and similar
261 tasks.
262
263 ## Version 2.3.3.1 - Oct 05, 2017 ##
264
265 * Fixed an issue causing input files to be skipped when running
266 multi-threaded alignment
267 * Fixed an issue causing the first character of a read name to be
268 dropped while parsing reads split across multiple input files
269
270 ## Version 2.3.3 - Sep 09, 2017 ##
271
272 From this release forward prepackaged bowtie2 binaries are now
273 statically linked to the zlib compression library and, the recommended
274 threading library, TBB. Users who rely on prepackaged builds are
275 no longer required to have these packages pre-installed. As a result
276 of the aforementioned changes legacy packages have been discontinued.
277
278 * bowtie2-build now supports gzip-compressed FASTA inputs
279 * New --xeq parameter for bowtie2 disambiguates the 'M' CIGAR
280 flag. When specified, matches are indicated with the '=' operation and
281 mismatches with 'X'
282 * Fixed a possible infinite loop during parallel index building due
283 to the compiler optimizing away a loop condition
284 * Added --soft-clipped-unmapped-tlen parameter for bowtie2 that
285 ignores soft-clipped bases when calculating template length (TLEN)
286 * Added support for multi-line sequences in FASTA read inputs
287 * Expanded explanation of MD:Z field in manual
288 * Fixed a crashing bug when output is redirected to a pipe
289 * Fixed ambiguity in the SEED alignment policy that sometimes caused -N parameter to be ignored
290
291 ## Version 2.3.2 - May 05, 2017 ##
292
293 * Added support for interleaved paired-end FASTQ inputs
294 (--interleaved)
295 * Now reports MREVERSE SAM flag for unaligned end when only one
296 end of a pair aligns
297 * Fixed issue where first character of some read names was
298 omitted from SAM output when using tabbed input formats
299 * Added --sam-no-qname-trunc option, which causes entire read
300 name, including spaces, to be written to SAM output. This
301 violates SAM specification, but can be useful in applications
302 that immediately postprocess the SAM.
303 * Fixed compilation error caused by pointer comparison issue
304 in aligner_result.cpp
305 * Removed termcap and readline dependencies introduced in v2.3.1
306 * Fixed compilation issues caused by gzbuffer function when
307 compiling with zlib v1.2.3.5 and earlier. Users compiling against
308 these libraries will use the zlib default buffer size of 8Kb
309 when decompressing read files.
310 * Fixed issue that would cause Bowtie 2 hang when aligning FASTA
311 inputs with more than one thread
312
313 ## Version 2.3.1 - Mar 04, 2017 ##
314
315 Please note that as of this release Bowtie 2 now has dependencies on zlib and readline libraries.
316 Make sure that all dependencies are met before attempting to build from source.
317
318 * Added native support for gzipped read files. The wrapper
319 script is no longer responsible for decompression. This
320 simplifies the wrapper and improves speed and thread scalability
321 for gzipped inputs.
322 * Fixed a bug that caused `bowtie2-build` to crash when the
323 first FASTA sequence contains all Ns.
324 * Add support for interleaved FASTQ format (`--interleaved`)
325 * Empty FASTQ inputs would yield an error in Bowtie 2 2.3.0,
326 whereas previous versions would simply align 0 reads and report
327 the SAM header as usual. This version returns to the pre-2.3.0
328 behavior, resolving a compatibility issue between TopHat2 and
329 Bowtie 2 2.3.0.
330 * Fixed a bug whereby combining `--un-conc*` with `-k` or `-a`
331 would cause `bowtie2` to print duplicate reads in one or both
332 of the `--un-conc*` output files, causing the ends to be
333 misaligned.
334 * The default `--score-min` for `--local` mode is now `G,20,8`.
335 That was the stated default in the documentation for a while,
336 but the actual default was `G,0,10` for many versions. Now the
337 default matches the documentation and, we find, yields more
338 accurate alignments than `G,0,10`
339
340 ## Version 2.3.0 - Dec 13, 2016 ##
341
342 * Code related to read parsing was completely rewritten to improve
343 scalability to many threads. In short, the critical section is
344 simpler and parses input reads in batches rather than one at a time.
345 The improvement applies to all read formats.
346 * TBB is now the default threading library. We consistently found TBB to give
347 superior thread scaling. It is widely available and widely installed.
348 That said, we are also preserving a "legacy" version of Bowtie that,
349 like previous releases, does not use TBB. To compile Bowtie source
350 in legacy mode use `NO_TBB=1`. To
351 use legacy binaries, download the appropriate binary archive with
352 "legacy" in the name.
353 * Bowtie now uses a queue-based lock rather
354 than a spin or heavyweight lock. We find this gives superior thread
355 scaling; we saw an order-of-magnitude throughput improvements at
356 120 threads in one experiment, for example.
357 * Unnecessary thread synchronization removed
358 * Fixed issue with parsing FASTA records with greater-than symbol in the name
359 * Now detects and reports inconsistencies between `--score-min` and `--ma`
360 * Changed default for `--bmaxdivn` to yield better memory footprint and running time when building an index with many threads
361
362 ## Version 2.2.9 - Apr 22, 2016 ##
363
364 * Fixed the multiple threads issue for the bowtie2-build.
365 * Fixed a TBB related build issue impacting TBB v4.4.
366
367 ## Version 2.2.8 - Mar 10, 2016 ##
368
369 * Various website updates.
370 * Fixed the bowtie2-build issue that made TBB compilation fail.
371 * Fixed the static build for Win32 platform.
372
373
374 ## Version 2.2.7 - Feb 10, 2016 ##
375
376 * Added a parallel index build option: bowtie2-build --threads <# threads>.
377 * Fixed an issue whereby IUPAC codes (other than A/C/G/T/N) in reads were
378 converted to As. Now all non-A/C/G/T characters in reads become Ns.
379 * Fixed some compilation issues, including for the Intel C++ Compiler.
380 * Removed debugging code that could impede performance for many alignment
381 threads.
382 * Fixed a few typos in documentation.
383
384 ## Version 2.2.6 - Jul 22, 2015 ##
385
386 * Switched to a stable sort to avoid some potential reproducibility confusions.
387 * Added 'install' target for *nix platforms.
388 * Added the Intel TBB option which provides in most situations a better performance
389 output. TBB is not present by default in the current build but can be added
390 by compiling the source code with WITH_TBB=1 option.
391 * Fixed a bug that caused seed lenght to be dependent of the -L and -N parameters order.
392 * Fixed a bug that caused --local followed by -N to reset seed lenght to 22 which is
393 actually the default value for global.
394 * Enable compilation on FreeBSD and clang, although gmake port is still required.
395 * Fixed an issue that made bowtie2 compilation process to fail on Snow Leopard.
396
397
398 ## Version 2.2.5 - Mar 9, 2015 ##
399
400 * Fixed some situations where incorrectly we could detect a Mavericks platform.
401 * Fixed some manual issues including some HTML bad formating.
402 * Make sure the wrapper correctly identifies the platform under OSX.
403 * Fixed --rg/--rg-id options where included spaces were incorrectly treated.
404 * Various documentation fixes added by contributors.
405 * Fixed the incorrect behavior where parameter file names may contain spaces.
406 * Fixed bugs related with the presence of spaces in the path where bowtie binaries are stored.
407 * Improved exception handling for missformated quality values.
408 * Improved redundancy checks by correctly account for soft clipping.
409
410 ## Version 2.2.4 - Oct 22, 2014 ##
411
412 * Fixed a Mavericks OSX specific bug caused by some linkage ambiguities.
413 * Added lz4 compression option for the wrapper.
414 * Fixed the vanishing --no-unal help line.
415 * Added the static linkage for MinGW builds.
416 * Added extra seed-hit output.
417 * Fixed missing 0-length read in fastq --passthrough output.
418 * Fixed an issue that would cause different output in -a mode depending on random seed.
419
420 ## Version 2.2.3 - May 30, 2014 ##
421
422 * Fixed a bug that made loading an index into memory crash sometimes.
423 * Fixed a silent failure to warn the user in case the -x option is missing.
424 * Updated al, un, al-conc and un-conc options to avoid confusion in cases
425 where the user does not provide a base file name.
426 * Fixed a spurious assert that made bowtie2-inspect debug fail.
427
428 ## Version 2.2.2 - April 10, 2014 ##
429
430 * Improved performance for cases where the reference contains ambiguous
431 or masked nucleobases represented by Ns.
432
433 ## Version 2.2.1 - February 28, 2014 ##
434
435 * Improved way in which index files are loaded for alignment. Should fix
436 efficiency problems on some filesystems.
437 * Fixed a bug that made older systems unable to correctly deal with bowtie
438 relative symbolic links.
439 * Fixed a bug that, for very big indexes, could determine to determine file
440 offsets correctly.
441 * Fixed a bug where using --no-unal option incorrectly suppressed
442 --un/--un-conc output.
443 * Dropped a perl dependency that could cause problems on old systems.
444 * Added --no-1mm-upfront option and clarified documentation for parameters
445 governing the multiseed heuristic.
446
447 ## Version 2.2.0 - February 10, 2014 ##
448
449 * Improved index querying efficiency using "population count" instructions
450 available since SSE4.2.
451 * Added support for large and small indexes, removing 4-billion-nucleotide
452 barrier. Bowtie 2 can now be used with reference genomes of any size.
453 * Fixed bug that could cause bowtie2-build to crash when reference length
454 is close to 4 billion.
455 * Fixed issue in bowtie2-inspect that caused -e mode not to output
456 nucleotides properly.
457 * Added a CL: string to the @PG SAM header to preserve information about
458 the aligner binary and paramteres.
459 * No longer releasing 32-bit binaries. Simplified manual and Makefile
460 accordingly.
461 * Credits to the Intel(r) enabling team for performance optimizations
462 included in this release. Thank you!
463 * Phased out CygWin support.
464 * Added the .bat generation for Windows.
465 * Fixed some issues with some uncommon chars in fasta files.
466 * Fixed wrappers so bowtie can now be used with symlinks.
467
468 Bowtie 2 on GitHub - February 4, 2014
469 * Bowtie 2 source now lives in a public GitHub repository:
470 https://github.com/BenLangmead/bowtie2.
471
472 ## Version 2.1.0 - February 21, 2013 ##
473
474 * Improved multithreading support so that Bowtie 2 now uses native Windows
475 threads when compiled on Windows and uses a faster mutex. Threading
476 performance should improve on all platforms.
477 * Improved support for building 64-bit binaries for Windows x64 platforms.
478 * Bowtie is using a spinlocking mechanism by default.
479 * Test option --nospin is no longer available. However bowtie2 can always
480 be recompiled with EXTRA_FLAGS="-DNO_SPINLOCK" in order to drop the
481 default spinlock usage.
482
483 ## Version 2.0.6 - January 27, 2013 ##
484
485 * Fixed issue whereby spurious output would be written in --no-unal mode.
486 * Fixed issue whereby multiple input files combined with --reorder would
487 cause truncated output and a memory spike.
488 * Fixed spinlock datatype for Win64 API (LLP64 data model) which made it
489 crash when compiled under Windows 7 x64.
490 * Fixed bowtie2 wrapper to handle filename/paths operations in a more
491 platform independent manner.
492 * Added pthread as a default library option under cygwin, and pthreadGC
493 for MinGW.
494 * Fixed some minor issues that made MinGW compilation fail.
495
496 ## Version 2.0.5 - January 4, 2013 ##
497
498 * Fixed an issue that would cause excessive memory allocation when aligning
499 to very repetitive genomes.
500 * Fixed an issue that would cause a pseudo-randomness-related assert to be
501 thrown in debug mode under rare circumstances.
502 * When bowtie2-build fails, it will now delete index files created so far so
503 that invalid index files don't linger.
504 * Tokenizer no longer has limit of 10,000 tokens, which was a problem for
505 users trying to index a very large number of FASTA files.
506 * Updated manual's discussion of the -I and -X options to mention that
507 setting them farther apart makes Bowtie 2 slower.
508 * Renamed COPYING to LICENSE and created a README to be GitHub-friendly.
509
510 ## Version 2.0.4 - December 17, 2012 ##
511
512 * Fixed issue whereby --un, --al, --un-conc and --al-conc options would
513 incorrectly suppress SAM output.
514 * Fixed minor command-line parsing issue in wrapper script.
515 * Fixed issue on Windows where wrapper script would fail to find
516 bowtie2-align.exe binary.
517 * Updated some of the index-building scripts and documentation.
518 * Updated author's contact info in usage message.
519
520 ## Version 2.0.3 - December 14, 2012 ##
521
522 * Fixed thread safely issues that could cause crashes with a large number of
523 threads. Thanks to John O'Neill for identifying these issues.
524 * Fixed some problems with pseudo-random number generation that could cause
525 unequal distribution of alignments across equally good candidate loci.
526 * The --un, --al, --un-conc, and --al-conc options (and their compressed
527 analogs) are all much faster now, making it less likely that they become
528 the bottleneck when Bowtie 2 is run with large -p.
529 * Fixed issue with innaccurate mapping qualities, XS:i, and YS:i flags when
530 --no-mixed and --no-discordant are specified at the same time.
531 * Fixed some compiler warnings and errors when using clang++ to compile.
532 * Fixed race condition in bowtie2 script when named pipes are used.
533 * Added more discussion of whitespace in read names to manual.
534
535 ## Version 2.0.2 - October 31, 2012 ##
536
537 * Fixes a couple small issues pointed out to me immediately after 2.0.1
538 release
539 * Mac binaries now built on 10.6 in order to be forward-compatible with more
540 Mac OS versions
541 * Small changes to source to make it compile with gcc versions up to 4.7
542 without warnings
543
544 ## Version 2.0.1 - October 31, 2012 ##
545
546 * First non-beta release.
547 * Fixed an issue that would cause Bowtie 2 to use excessive amounts of
548 memory for closely-matching and highly repetitive reads under some
549 circumstances.
550 * Fixed a bug in --mm mode that would fail to report when an index file
551 could not be memory-mapped.
552 * Added --non-deterministic option, which better matches how some users
553 expect the pseudo-random generator inside Bowtie 2 to work. Normally, if
554 you give the same read (same name, sequence and qualities) and --seed, you
555 get the same answer. --non-deterministic breaks that guarantee. This can
556 be more appropriate for datasets where the input contains many identical
557 reads (same name, same sequence, same qualities).
558 * Fixed a bug in bowtie2-build would yield corrupt index files when memory
559 settings were adjusted in the middle of indexing.
560 * Clarified in manual that --un (or --un-*) options print reads exactly as
561 they appeared in the input, and that they are not necessarily written in
562 the same order as they appeared in the input.
563 * Fixed issue whereby wrapper would incorrectly interpret arguments with
564 --al as a prefix (e.g. --all) as --al.
565 * Added option --omit-sec-seq, which causes Bowtie 2 to set SEQ and QUAL
566 fields to "*" for secondary alignments.
567
568 ## Version 2.0.0-beta7 - July 9, 2012 ##
569
570 * Fixed an issue in how Bowtie 2 aligns longer reads in --local mode. Some
571 alignments were incorrectly curtailed on the left-hand side.
572 * Fixed issue --un (or --un-*) would fail to output unaligned reads when
573 --no-unal was also specified.
574 * Fixed issue whereby --un-* were significantly slowing down Bowtie 2 when
575 -p was set greater than 1.
576 * Fixed issue that would could cause hangs in -a mode or when -k is set
577 high.
578 * Fixed issue whereby the SAM FLAGS field could be set incorrectly for
579 secondary paired-end alignments with -a or -k > 1.
580 * When input reads are unpaired, Bowtie 2 no longer removes the trailing /1
581 or /2 from the read name.
582 * -M option is now deprecated. It will be removed in subsequent versions.
583 What used to be called -M mode is still the default mode, and -k and -a
584 are still there alternatives to the default mode, but adjusting the -M
585 setting is deprecated. Use the -D and -R options to adjust the effort
586 expended to find valid alignments.
587 * Gaps are now left-aligned in a manner similar to BWA and other tools.
588 * Fixed issue whereby wrapper script would not pass on exitlevel correctly,
589 sometimes spuriously hiding non-0 exitlevel.
590 * Added documentation for YT:Z to manual.
591 * Fixed documentation describing how Bowtie 2 searches for an index given an
592 index basename.
593 * Fixed inconsistent documentation for the default value of the -i parameter
594
595 ## Version 2.0.0-beta6 - May 7, 2012 ##
596
597 * Bowtie 2 now handles longer reads in a more memory-economical fashion,
598 which should prevent many out-of-memory issues for longer reads.
599 * Error message now produced when -L is set greater than 32.
600 * Added a warning message to warn when bowtie2-align binary is being run
601 directly, rather than via the wrapper. Some functionality is provided by
602 the wrapper, so Bowtie 2 should always be run via the bowtie2 executable
603 rather than bowtie2-align.
604 * Fixed an occasional crashing bug that was usually caused by setting the
605 seed length relatively short.
606 * Fixed an issue whereby the FLAG, RNEXT and PNEXT fields were incorrect for
607 some paired-end alignments. Specifically, this affected paired-end
608 alignments where both mates aligned and one or both mates aligned
609 non-uniquely.
610 * Fixed issue whereby compressed input would sometimes be mishandled.
611 * Renamed --sam-* options to omit the "sam-" prefix for brevity. The old
612 option names will also work.
613 * Added --no-unal option to suppress SAM records corresponding to unaligned
614 reads, i.e., records where FLAG field has 0x4 set.
615 * Added --rg-id option and enhanced the documentation for both --rg-id and
616 --rg. Users were confused by the need to specify --rg "ID:(something)" in
617 order for the @RG line to be printed; hopefully this is clearer now.
618 * Index updates: indexes linked to in the right-hand sidebar have been
619 updated to include the unplaced contigs appearing in the UCSC "random"
620 FASTA files. This makes the indexes more complete. Also, an index for
621 the latest mouse assembly, mm10 (AKA "GRCm38") has been added.
622
623 ## Version 2.0.0-beta5 - December 14, 2011 ##
624
625 * Added --un, --al, --un-conc, and --al-conc options that write unpaired
626 and/or paired-end reads to files depending on whether they align at least
627 once or fail to align.
628 * Added --reorder option. When enabled, the order of the SAM records output
629 by Bowtie 2 will match the order of the input reads even when -p is set
630 greater than 1. This is disabled by default; enabling it makes Bowtie 2
631 somewhat slower and use somewhat more memory when -p is set greater than
632 1.
633 * Changed the default --score-min in --local mode to G,20,8. This ought to
634 improve sensitivity and accuracy in many cases.
635 * Improved error reporting.
636 * Fixed some minor documentation issues.
637 * Note: I am aware of an issue whereby longer reads (>10,000 bp) drive the
638 memory footprint way up and often cause an out-of-memory exception. This
639 will be fixed in a future version.
640
641 ## Version 2.0.0-beta4 - December 5, 2011 ##
642
643 * Accuracy improvements.
644 * Speed improvements in some situations.
645 * Fixed a handful of crashing bugs.
646 * Fixed some documentation bugs.
647 * Fixed bug whereby --version worked incorrectly.
648 * Fixed formatting bug with MD:Z optional field that would sometimes fail to
649 follow a mismatch with a number.
650 * Added -D option for controlling the maximum number of seed extensions that
651 can fail in a row before we move on. This option or something like it
652 will eventually replace the argument to -M.
653 * Added -R option to control maximum number of times re-seeding is attempted
654 for a read with repetitive seeds.
655 * Changed default to --no-dovetail. Specifying --dovetail turns it back on.
656 * Added second argument for --mp option so that user can set maximum and
657 minimum mismatch penalties at once. Also tweaked the formula for
658 calculating the quality-aware mismatch penalty.
659
660 ## Version 2.0.0-beta3 - November 1, 2011 ##
661
662 * Accuracy improvements.
663 * Speed improvements in some situations.
664 * Fixed a handful of crashing bugs.
665 * Fixed a bug whereby number of repetitively aligned reads could be
666 misreported in the summary output.
667
668 ## Version 2.0.0-beta2 - October 16, 2011 ##
669
670 * Added manual, both included in the download package and on the website.
671 The website will always have the manual for the latest version.
672 * Added Linux 32-bit and 64-bit binary packages. Mac OS X packages to come.
673 Still working on a Windows package.
674 * Fixed a bug that led to crashes when seed-alignment result memory was
675 exhausted.
676 * Changed the --end-to-end mode --score-min default to be less permissive.
677 The previous threshold seemed to be having an adverse effect on accuracy,
678 though the fix implemented in this version comes at the expense of some
679 sensitivity.
680 * Changed the --end-to-end mode -M default to be lower by 2 notches. This
681 offsets any detrimental effect that the previous change would have had on
682 speed, without a large adverse impact on accuracy. As always, setting -M
683 higher will yield still greater accuracy at the expense of speed.
684
685 ## Version 2.0.0-beta1 - September 22, 2011 ##
686
687 * First public release.
688 * Caveats: as of now, the manual is incomplete, there's no tutorial, and no
689 example genome or example reads. All these will be fixed in upcoming
690 releases.
691 * Only a source package is currently available. Platform-specific binaries
692 will be included in future releases.