minibwa/minibwa.1 at dist · nh13/minibwa

Repository navigation

History

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

.TH minibwa 1 "6 August 2026" "minibwa-0.7 (r421)" "Bioinformatics tools"

.SH NAME

.PP

minibwa - short/long-read alignment against a reference genome

.SH SYNOPSIS

* Indexing a genome with the libsais algorithm (OpenMP required for multi-threading):

.RS 4

minibwa index

.RB [ -t

.IR nThreads ]

.I ref.fa

.RI [ idxPrefix ]

.RE

* Indexing a genome in low memory with the BWT-SW algorithm:

.RS 4

minibwa index

.B -l

.I ref.fa

.RI [ idxPrefix ]

.RE

* Aligning short paired-end reads to an indexed genome and outputting SAM:

.RS 4

minibwa map

.RB [ -t

.IR nThreads ]

.I idxPrefix

.I read1.fq.gz

.I read2.fq.gz

.I output.sam

.RE

* Indexing and aligning bisulfite sequencing reads:

.RS 4

minibwa index

.B --meth

.RB [ -t

.IR nThreads ]

.I ref.fa

.RI [ idxPrefix ]

.br

minibwa map

.B --meth

.RB [ -t

.IR nThreads ]

.I idxPrefix

.I read1.fq.gz

.I read2.fq.gz

.I output.sam

.RE

.SH DESCRIPTION

Minibwa aligns short reads against a reference genome. It is the successor of

bwa-mem with a different algorithm. Minibwa is over three times as fast as the

original bwa-mem and twice as fast as bwa-mem2 at comparable accuracy. While

minibwa works with accurate long reads, minimap2 is more robust under high

error rate.

Minibwa is a hybrid of bwa-mem and minimap2: it indexes the genome with

Burrow-Wheeler Transform (BWT), finds variable-length seeds like bwa-mem, and

performs chaining and SIMD-based nucleotide alignment with the minimap2

algorithm. Minibwa speeds up bwa-mem2 further with additional prefetch for

seeding, new heuristics to skip unnecessary mate rescue and reduced effort in

highly repetitive regions where reads would often be wrongly mapped due to

structural changes anyway.

.SH THE INDEX COMMAND

.SS Summary

.B minibwa index

.RB [ -l ]

.RB [ -t

.IR nThreads ]

.RB [ --meth ]

.I ref.fa

.RI [ idxPrefix ]

Index a genome in the FASTA format. Generate

.IR idxPrefix .l2b

for the genome sequence and

.IR idxPrefix .mbw

for the FM-index. Minibwa uses the libsais algorithm for BWT construction by

default. This algorithm is fast but requires

.RI 18 N

bytes where

.I N

is the length of the genome. Use

.B -l

to enable the BWT-SW algorithm which is slower but uses much less memory.

.SS Options

.TP 10

.B -l

Use the old low-memory algorithm for BWT construction. Minibwa uses the libsais

algorithm by default which uses over 50 GB for a human genome.

.TP

.BI -t \ INT

Number of threads for the libsais algorithm [1]. Minibwa needs to be compiled

with OpenMP to enable this option.

.TP

.B --meth

Generate an index for aligning bisulfite sequencing reads. Minibwa will

additionally generate

.IR idxPrefix .meth.mbw

which combines C-to-T and G-to-A converted reference genomes.

.SH THE MAP COMMAND

.SS Summary

.B minibwa map

.RB [ -x

.IR preset ]

.RB [ -t

.IR nThreads ]

.I idxPrefix

.I read.fq.gz

.RI [ read2.fq.gz ]

Map reads to an index genome. Minibwa automatically detects

interleaved read pairs and like bwa-mem and minimap2, allowing mixed single-end

and paired-ened reads as input.

.SS Common Options

.TP 10

.B -f

Output alignment in the PAF format (SAM by default)

.TP

.BI -t \ INT

Number of worker CPU threads [1]. Minibwa additionally uses

.RI min(2, \fIINT )

threads for I/O.

.TP

.BI -l \ NUM

In the adaptive mode, treat reads shorter than

.I NUM

as short reads [325]

.TP

.BI -R \ STR

SAM read group line in a format like '@RG\\tID:foo\\tSM:bar' []

.TP

.BI -b \ STR

Output a base alignment tag:

.BR cs,

.B ds

.B MD

[]. Only one tag is allowed.

.TP

.B --hic

Map Hi-C reads. Equivalent to option

.BR -5P .

.TP

.B --meth

Map

.B directional

bisulfite sequencing (BS-seq) data. Require an index built with

.BR --meth .

In this mode, an unmethylated C on read 1 is expected to be converted to T and

an unmethylated G on read 2 converted to A. Minibwa does not support

undirectional BS-seq.

In the SAM output, each alignment carries the Bismark

.BR XR:Z ,

.B XG:Z

and

.B XM:Z

tags; see

.BR --meth-tags .

.TP

.BR --meth-tags = \fISTR

With

.BR --meth ,

select which of the three Bismark tags to emit [all].

.I STR

.BR all ,

.BR none ,

a comma-separated list of tag names to emit, or

.B ^

followed by such a list of tag names to omit \(em for example

.B XR,XG

and

.B ^XM

both mean "emit XR and XG". Tag names are case-insensitive, and the list is a

set: order and repetition do not matter. The two list forms cannot be combined:

.B XR,^XM

is an error rather than a guess between "only XR" and "everything but XM". The

separator is a comma and never a space. A malformed

.I STR

is an error; minibwa never falls back to the default.

.IP

.B XM

is the reason this option exists. It is a per-base call string as long as the

read, so it dominates the size of the methylation output, while

.B XR

and

.B XG

are two-character constants. Dropping it is the difference between an alignment

file that carries per-base calls and one that does not; it does not change the

alignments themselves, and it costs no mapping time either way.

The default is

.B all

because the tags are cheap to keep and expensive to miss: consumers in the

Bismark family, such as

.BR bismark_methylation_extractor ,

methylKit, methtuple, DMRfinder and epialleleR, read

.B XM

as an input contract and cannot rebuild it from the alignment, so silently

dropping it yields wrong methylation calls. Downstream tools that recompute the

methylation context from the reference FASTA never read

.B XM

and can safely ask for

.BR ^XM .

.SS Mapping Options

.TP 10

.BI -k \ INT

Minimum seed length [19]. Like bwa-mem, minibwa uses variable-length seeds.

.TP

.BI -c \ NUM

Maximum seed occurrence [250]. Minibwa samples up to

.I NUM

occurrences if there are too many.

.TP

.BI -g \ NUM

Stop chain enlongation if there are no minimizers within

.IR NUM -bp

[5000]. Also used for extending from the first and last anchors during base

alignment.

.TP

.BI -w \ NUM

Bandwidth for initial chaining and base alignment [500].

.TP

.BI -W \ NUM

Bandwidth for secondary chaining through long gaps [20000].

.TP

.BI -m \ INT

Minimum chaining score [25]

.TP

.BI -p \ FLOAT

Minimum secondary-to-primary score ratio [0.5]

.TP

.BI -N \ INT

Retain at most

.I INT

secondary alignments [101]

.TP

.B --chain-only

Perform chaining only without base alignment. Force

.B -fP

to disable pairing and output PAF.

.TP

.BR --long = yes | no

Whether to use the long-read heuristic for all reads

.TP

.BR --adap = yes | no

Whether to enable the adaptive mode, which is the default behavior with the

.B adap

preset. In the adaptive mode, minibwa smoothly adjusts

.BR -g / -w / -W / -N / -m / -s

based on the individual read length, using formula

.IR x1 -( x1 - x0 )*pow(2,-(max( L ,100)-100)/(2000-100)),

where

.I L

is the read length,

.I x1

is the value for long reads that are fixed, and

.I x0

is the value for short reads that can be changed on the command line. For a

read longer than

.BR -l ,

minibwa additionally enables anchor filtering and mixes the short- and

long-read formulae for calculating the mapping quality.

.TP

.BI -x \ STR

Preset:

.BR sr ,

.B lr

.B adap

< 8000 /div>

.RB [ adap ].

If unsure, use the default

.B adap

which works well for both short and long reads.

.RS

.TP 6

.B sr

For short-read alignment. It may give fragmented alignment for typical long reads.

.TP

.B lr

For long-read alignment

.RB ( --pe=no

.B --long -m40 -s50 -w500 -g5k -N5

.BR -K500m ).

It may miss highly repetitive alignments for typical short reads.

.TP

.B adap

For mixed short and long reads but optimized for short reads. It is equivalent to

.RB ' -xsr

.BR --adap=yes '.

.RE

.SS Alignment Options

.TP 10

.BI -A \ INT

Match score [2]

.TP

.BI -B \ INT

Mismatch penalty [8]

.TP

.B -O \fIINT1\fR[,\fIINT2\fR]

Dual gap open penalty [12,23]. If

.I INT2

is not specified, it is set to

.IR INT1 .

.TP

.B -E \fIINT1\fR[,\fIINT2\fR]

Dual gap extension penalty [2,1]. If

.I INT2

is not specified, it is set to

.IR INT1 .

A gap of length

.I k

costs

.RI min( O1 + k * E1 , O2 + k * E2 ).

.TP

.B -L \fIINT1\fR[,\fIINT2\fR]

5'- and 3'-end soft-clip penalty [0,0]. After the DP extension, the

extend-to-end alignment is preferred over a clipped local-best

alignment iff

.RI ( mqe + end_bonus + L >= max ),

where

.I L

is the relevant

.B -L

value (

.I INT1

for 5' /

.I INT2

for 3'). If

.I INT2

is not specified, it is set to

.IR INT1 .

Higher

.I L

makes the aligner more willing to extend through a noisy read tail

rather than soft-clip it. Mirrors bwa-mem's

.BR -L

flag; default 0,0 preserves historical minibwa behavior (bwa-mem

defaults to 5,5).

.SS Paired-end Options

.TP 10

.B -P

Skip read pairing and mate rescue

.TP

.BR --rescue = \fIINT

Mate rescue for up to

.I INT

candidates [10]

.TP

.BR --pe = yes | no

Whether to enable the paired-end mode. Enabled by default with the

.B adap

.B sr

presets. Disabled with the

.B lr

preset.

.TP

.B -I \fIINT1\fR[,\fIINT2\fR[,\fIINT3\fR[,\fIINT4\fR]]]

Mean, standard deviation, max and min of the expected insert size distribution [inferred].

If missing,

.IR INT2 =round( INT1 /10),

.IR INT3 = INT1 + INT2 *4,

and

.IR INT4 =min( INT1 - INT2 *4,1).

.SS Input/Output Options

.TP 10

.BI -o \ FILE

Output file name [stdout]

.TP

.B -u

Don't output unmapped reads

.TP

.BR --outn = \fINUM

Output up to

.RI min( INT ,{\fB-N\fR})

secondary alignments [0]. Primary and supplementary alignments are always outputted.

.TP

.BR --outs = \fIFLOAT

Output a secondary hit if its score is above

.I FLOAT

times the primary alignment score [0.8]. This option also controls the XA tag.

.TP

.BR --xa = \fINUM

If there are

.RI >= NUM

secondary hits of score above

.RB { --outs }

times the primary alignment score, output all of them to the XA tag [5]

.TP

.B -y

Copy FASTA/Q comments to output

.TP

.B -Y

In the SAM output, use soft clipping for supplementary alignments

.TP

.BI -H \ STR

.I STR

starts with @, insert it to the SAM header; otherwise, treat

.I STR

as a file name and insert lines in this file to the SAM header []

.TP

.B -5

In the SAM output, if a read has supplementary alignments, take the alignment

with the smallest query (not genomic) position as primary. This mode, combined with

.BR -P ,

is commonly used for mapping Hi-C reads.

.TP

.B -K \fINUM1\fR[,\fINUM2\fR]

Process

.IR NUM1 - NUM2

bp of query sequences in a batch [100m,1g]. Given enough reads in the input, a

batch is at least

.I NUM1

in size. If there are less than 40000 sequences in the batch, continue to read

until the size reaches

.IR NUM2 .

.TP

.BR --mmap [= lite ]

Load the index by mapping the index files to RAM. With this option, the

first index loading time will be slower but the following index loading will be

faster if the index remains in RAM. The optional

.B lite

argument further reduces hot loading time but may greatly increase cold loading

time sometimes. With

.BR --mmap ,

multiple minibwa processes on the same machine can also use the same index

without duplicating it in memory. Mapping may be slower depending on the

system.

.TP

.B --version

Print version number

.SH OUTPUT FORMATS

.PP

Minibwa outputs the SAM format by default. Options

.B -f

enables the PAF output. PAF is a TAB-delimited text format with each line

consisting of at least 12 fields as are described in the following table:

.TS

center box;

cb | cb | cb

r | c | l .

Col Type Description

1 string Read name

2 int Read length

3 int Read start coordinate (0-based)

4 int Read end coordinate (0-based)

5 char `+' for forward strand; `-' for reverse

6 string Contig name

7 int Contig length

8 int Contig start coordinate on the original strand

9 int Contig end coordinate on the original strand

10 int Number of matching bases

11 int Number of bases in alignment

12 int Mapping quality (0-255 with 255 for missing)

.TE

.PP

Both SAM and PAF may have addtional fields in typed

key-value format. Minibwa may output the following tags:

.TS

center box;

cb | cb | cb

r | c | l .

Tag Type Description

NM i Edit distance

AS i Alignment score from dynamic programming

ms i Score of the maximum scoring segment

md i Max score minus secondary max score

XA Z Secondary hits in (ctg,strandPos,CIGAR,NM;)+

MC Z Mate CIGAR

MQ i Mate mapping quality

cg Z CIGAR (PAF only)

MD Z Difference string in the SAM spec

cs Z Difference string

.TE

.SH LIMITATIONS

.TP 2

Minibwa does not work with noisy long reads or spliced RNA-seq reads.

.TP

Minibwa does not support undirectional bisulfite sequencing data.

.TP

Minibwa does not recognize alternate haplotypes.

FilesExpand file tree

minibwa.1

Latest commit

History

minibwa.1

File metadata and controls