Development manual for vsearch 2.32.0, built from the dev branch: it describes changes that are not released yet. The manual for the current release is published separately.

NAME

vsearch --sff_convert — convert an SFF file to fastq

SYNOPSIS

vsearch --sff_convert sfffile --fastqout filename [options]

DESCRIPTION

The vsearch command --sff_convert converts the reads stored in an SFF (Standard Flowgram Format) file to fastq. SFF is a binary format used by Roche 454 and early Ion Torrent PGM sequencing platforms (see vsearch-sff(5)).

The output fastq file is written to --fastqout. The quality encoding offset can be set with --fastq_asciiout (default 33, phred+33). Output quality scores are always clamped to the range set by --fastq_qminout and --fastq_qmaxout, whose defaults are 0 and 93: SFF quality values are written unchanged unless --fastq_qmaxout is lowered. Before version 2.32.0 the default was 41, and any higher SFF value was silently reduced to it.

While converting, vsearch reports the number of reads, the flows per read, the key sequence, the index type (when an index is present) and the minimal, average and maximal read lengths — to standard error, or to the log file when --log is used. A warning is emitted when the SFF index is missing, when the index is at an unusual position in the file, or when additional data follow the end of the SFF content.

Each SFF read stores clipping coordinates that indicate low-quality or adapter regions at the ends of the sequence. By default, no clipping is applied: the full sequence is written, with bases that would be clipped converted to lower case and the rest in upper case. Use --sff_clip to apply the clipping and write only the retained region in upper case.

OPTIONS

mandatory options

--fastqout filename: Write converted sequences to filename, in fastq format (see vsearch-fastq(5)).

core options

--fastq_asciiout 33|64: Specify the offset used as the basis for the fastq quality score when writing fastq output files. For example, an offset of 33 means that a quality value of 41 is represented by the 74th ASCII symbol (33 + 41 = 74), which is ‘J’. See ascii(7) for a view of the ASCII character set. The offset value is either 33 or 64, default is 33.
--fastq_qmaxout integer: Specify the maximum quality score used when writing fastq files. The default is the highest score the output offset can represent (93 with --fastq_asciiout 33, 62 with offset 64), so scores read from the input are written back unchanged. Before version 2.32.0 the default was 41, the usual maximum for Sanger/Illumina 1.8+ files, which silently reduced any higher score. Older formats may use a maximum quality score of 40. Two commands are exceptions and keep the old default of 41, because they generate the score they clamp instead of passing one through: --fasta2fastq, which has no input quality and uses this option as the value to write, and --fastq_mergepairs, which caps the computed posterior quality of a merged base.

For --fastq_mergepairs the offset in question is --fastq_ascii, not --fastq_asciiout: that command writes fastq but does not accept --fastq_asciiout, so a merged quality symbol carries the same offset the input was read with. The sum rule is stated against --fastq_ascii there, and against --fastq_asciiout everywhere else.

--fastq_qminout integer: Specify the minimum quality score used when writing fastq files. The offset (see --fastq_asciiout) plus the minimum score must be at least 33, so the value may be negative when the output offset is 64. The default is 0, which is usual for recent Sanger/Illumina 1.8+ files. Older formats may use scores between -5 and 2.

A negative minimum matters where the score passes through vsearch and may itself be negative: --fastq_convert on an offset-64 file read with a lowered --fastq_qmin writes the score back unchanged with --fastq_qminout -5, where the default 0 would raise it. It has no effect where vsearch computes the score it clamps, because such a score is derived from an error probability and is therefore never negative: --fastq_mergepairs, and --fastq_convert --fastq_solexa, which converts to the Phred scale before any output clamp applies.

--sff_clip: Clip sequences at both ends as indicated by the clipping coordinates stored in the SFF file (clip_qual_left, clip_qual_right, clip_adapter_left, clip_adapter_right). Without this option, no clipping is performed and bases that would have been clipped are written in lower case, while the remaining bases are in upper case.

secondary options

--label_suffix string: Add the suffix string to sequence headers when writing fasta or fastq files. For example, with --label_suffix ";status=healthy", sequence header ‘>seq1’ becomes ‘>seq1;status=healthy’.
--lengthout: Add a sequence length annotation (;length=integer) to each sequence header when writing fasta or fastq files.
--log filename: Write messages to filename. Messages include program version, start and finish times, elapsed time, amount of memory available, maximum amount of memory consumed, number of cores and command line options, and if need be, command-specific informational messages, warnings, and errors. Messages are written in addition to the standard error, not instead of it; use --quiet to silence the standard error copy (see the --quiet entry for what it spares).
--no_progress: Suppress the gradually increasing progress indicator normally written to the standard error stderr(3).
--quiet: Suppress messages to the standard output stdout(3) and standard error stderr(3), except for warnings and error messages.
--relabel string: Replace sequence headers with the prefix string and a ticker (1, 2, 3, etc.). For example, with --relabel "cluster:", the first sequence header becomes ‘>cluster:1’, the second sequence header becomes ‘>cluster:2’, and so on. To retain annotations, use their corresponding options (--lengthout, --eeout, and --sizeout). Use --relabel_keep to also retain old sequence identifiers.
--relabel_keep: Retain old sequence identifiers by including them at the end of the new headers, after a space.
--relabel_md5: Replace each sequence header with the MD5 digest derived from the sequence itself. The sequence is converted to upper case, and each ‘U’ is replaced with a ‘T’ before computation of the digest. The MD5 digest is a 128-bit value (16 bytes), represented using a string of 32 ASCII characters. Each pair of characters encodes an hexadecimal value, ranging from x00 to xff. See md5(3) for more details, and --relabel_sha1 for an alternative hashing algorithm. To retain annotations, use their corresponding options (--lengthout, --eeout, and --sizeout). Use --relabel_keep to also retain old sequence identifiers.
--relabel_self: Replace each sequence header with the sequence itself. To retain annotations, use their corresponding options (--lengthout, --eeout, and --sizeout). Use --relabel_keep to also retain old sequence identifiers.
--relabel_sha1: Replace each sequence header with the SHA1 digest derived from the sequence itself. The sequence is converted to upper case, and each ‘U’ is replaced with a ‘T’ before computation of the digest. The SHA1 digest is a 160-bit value (20 bytes), represented using a string of 40 ASCII characters. Each pair of characters encodes an hexadecimal value, ranging from x00 to xff. See sha1(3) for more details, and --relabel_md5 for an alternative hashing algorithm. To retain annotations, use their corresponding options (--lengthout, --eeout, and --sizeout). Use --relabel_keep to also retain old sequence identifiers.
--sample string: Add the given sample identifier string to sequence headers when writing fasta or fastq files. For instance, if string is ‘ABC’, the text ;sample=ABC will be added to the headers. string is silently truncated at the first ‘;’ or whitespace character (space, tab, newline, carriage return, vertical tab or form feed), so such characters should not be used in string. Other characters (alphabetical, numerical and punctuations) are accepted. When nothing is left after truncation — an empty string, or one starting with ‘;’ or a blank character — vsearch issues a warning and writes a bare ;sample= annotation.
--sizeout: Add abundance annotations to sequence headers when writing fasta or fastq files. Add the pattern ;size=integer. Existing ;size= annotations are reported unchanged; entries without one receive ;size=1. For this command --sizein is not needed: abundance annotations are always parsed from the input headers.

ignored options

--threads positive non-null integer: Command is not multithreaded, option has no effect (a warning is printed when more than one thread is requested).

EXAMPLES

Convert an SFF file to fastq using the default phred+33 encoding:

vsearch \
    --sff_convert reads.sff \
    --fastqout reads.fastq

Convert and apply clipping coordinates from the SFF file:

vsearch \
    --sff_convert reads.sff \
    --sff_clip \
    --fastqout reads_clipped.fastq

Convert with phred+64 encoding and restrict quality scores to 0–40:

vsearch \
    --sff_convert reads.sff \
    --fastq_asciiout 64 \
    --fastq_qmaxout 40 \
    --fastqout reads.fastq

CITATION

Rognes T, Flouri T, Nichols B, Quince C, Mahé F. (2016) VSEARCH: a versatile open source tool for metagenomics. PeerJ 4:e2584 doi: 10.7717/peerj.2584

REPORTING BUGS

Submit suggestions and bug-reports at https://github.com/torognes/vsearch/issues, send a pull request on https://github.com/torognes/vsearch, or compose a friendly or curmudgeont e-mail to Torbjørn Rognes (torognes@ifi.uio.no).

AVAILABILITY

Source code and binaries are available at https://github.com/torognes/vsearch.

These manual pages are also published online at https://torognes.github.io/vsearch/.

COPYRIGHT

Contact: Torbjørn Rognes torognes@ifi.uio.no, Department of Informatics, University of Oslo, PO Box 1080 Blindern, NO-0316 Oslo, Norway

This software is dual-licensed and available under a choice of one of two licenses, either under the terms of the GNU General Public License version 3 or the BSD 2-Clause License.

GNU General Public License version 3

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.

The BSD 2-Clause License

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS “AS IS” AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

ACKNOWLEDGMENTS

We would like to thank the authors of the following projects for making their source code available:

vsearch includes code from Google’s CityHash project by Geoff Pike and Jyrki Alakuijala, providing some excellent hash functions available under a MIT license.
vsearch includes code derived from Tatusov and Lipman’s DUST program that is in the public domain.
vsearch includes public domain code written by Alexander Peslyak for the MD5 message digest algorithm.
vsearch includes public domain code written by Steve Reid and others for the SHA1 message digest algorithm.
vsearch binaries may include code from the zlib library, copyright Jean-Loup Gailly and Mark Adler.
vsearch binaries may include code from the bzip2 library, copyright Julian R. Seward.