vsearch

Versatile open-source tool for microbiome analysis

View on GitHub

NAME

vsearch \-\-fastx_uniques — merge strictly identical fasta or fastq sequences

SYNOPSIS

vsearch \-\-fastx_uniques fastxfile (\-\-fastaout | \-\-fastqout) filename [options]

DESCRIPTION

The vsearch command --fastx_uniques groups strictly identical sequences from a fasta or fastq file, like --derep_fulllength (see vsearch-derep_fulllength(1)). The unique sequences are written to --fastaout and/or --fastqout, sorted by decreasing abundance.

When the input is a fastq file, quality scores in the output correspond by default to the average error probability at each position across all grouped sequences (see vsearch-expected_error(7)). Use --fastq_qout_max to use the best (highest) quality score observed at each position instead.

This command is not multithreaded. The inverse operation is --rereplicate (see vsearch-rereplicate(1)).

See vsearch-fasta(5) and vsearch-fastq(5) for a description of the input formats.

OPTIONS

mandatory options

At least one of --fastaout or --fastqout must be specified.

core options

--fastaout filename
Write the dereplicated sequences to filename, in fasta format, sorted by decreasing abundance. Each unique sequence retains the header of the first occurrence in the input.

--fastqout filename
Write the dereplicated sequences to filename, in fastq format (see vsearch-fastq(5)), sorted by decreasing abundance. Quality scores correspond to the average error probability at each position across all grouped sequences, or to the best quality score if --fastq_qout_max is set.

--fastq_qout_max
Use the best (highest) quality score observed at each position when computing quality scores for fastq output, instead of averaging error probabilities.

--maxuniquesize positive integer
Discard sequences with a post-dereplication abundance greater than positive integer.

--minuniquesize positive integer
Discard sequences with a post-dereplication abundance smaller than positive integer.

--sizein
Use the abundance annotations present in sequence headers when reading fasta or fastq file. Search for the pattern [>@;]size=integer[;]. Entries without abundance annotations are silently assumed to be of size=1.

--sizeout
Add abundance annotations to sequence headers when writing fasta or fastq files. Add the pattern ;size=integer. If option --sizein is not used, abundance values are set to 1 for all entries. If --sizein is used, existing abundance annotations are simply reported to output files.

--strand plus|both
Check the plus strand only (default), or check both strands when comparing sequences.

--tabbedout filename
Write dereplication details to filename as a tab-separated file with one row per input sequence and 6 columns:

  1. original sequence label;
  2. output label (label of the first sequence in the cluster, possibly relabelled);
  3. cluster number (zero-based);
  4. sequence number within the cluster (zero-based);
  5. cluster size;
  6. original label of the first sequence in the cluster (before any relabelling).

--topn positive non-null integer
Write only the n first entries. The parameter n must be greater than zero, and is silently ignored if it is greater than the total number of entries.

--uc filename
Write results to filename in a tab-separated uclust-like format with 10 columns. Three record types are used per row: cluster seeds (S), hits (H), and cluster summaries (C). Columns are:

  1. record type (S, H, or C);
  2. cluster number (zero-based);
  3. centroid length (S), query length (H), or cluster size (C);
  4. percent identity with centroid (H), or * (S, C);
  5. match orientation + or - (H), or * (S, C);
  6. not used; always 0 (H) or * (S, C);
  7. not used; always 0 (H) or * (S, C);
  8. CIGAR alignment string (H), or * (S, C); see vsearch-cigar(5);
  9. query label (H), or centroid label (S, C);
  10. centroid label (H), or * (S, C).

secondary options

--bzip2_decompress
Specify that the input pipe is streaming data compressed using Huffman coding. See bzip2(1) for more details. This option is not needed when reading from a regular file compressed with bzip2.

--fasta_width positive integer
Set the maximal width of sequences when writing fasta files. Longer sequences are folded and written on several lines. Default width is 80 nucleotides. Set to zero (0) to suppress folding.

--fastq_ascii 33|64
Specify the offset used as the basis for the fastq quality score when reading fastq 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_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_qmax positive integer
Specify the maximal quality score accepted when reading fastq sequences. Stop with an error message if a quality score higher than the specified value is read. Accepted values range from 0 to 93 if the offset is 33 (see --fastq_ascii), or range from 0 to 62 if the offset is 64. The default is 41, which is usual for recent Sanger/Illumina 1.8+ files.

--fastq_qmaxout positive integer
Specify the maximum quality score used when writing fastq files. The default is 41, which is usual for recent Sanger/Illumina 1.8+ files. Older formats may use a maximum quality score of 40.

--fastq_qmin positive integer
Specify the minimal quality score accepted when reading fastq sequences. Stop with an error message if a quality score lower than the specified value is read. Accepted values range from 0 to 93 if the offset is 33 (see --fastq_ascii), or range from 0 to 62 if the 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.

--fastq_qminout positive integer
Specify the minimum quality score used when writing fastq files. The default is 0, which is usual for recent Sanger/Illumina 1.8+ files. Older formats may use scores between -5 and 2.

--gzip_decompress
Specify that the input pipe is streaming data compressed using Lempel-Ziv coding. See gzip(1) for more details. This option is not needed when reading from a regular file compressed with gzip.

--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.

--maxseqlength positive integer
Discard sequences longer than positive integer (50,000 nucleotides by default).

--minseqlength positive integer
Discard sequences shorter than positive integer (1 nucleotide by default).

--no_progress
Suppress the gradually increasing progress indicator normally written to the standard error stderr(3).

--notrunclabels
Retain whole sequence headers in output files. By default, vsearch truncates sequence headers at first space or tabulation. This option suppresses truncation.

--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.

--xee
Strip expected error (ee) annotations from sequence headers when writing fasta or fastq files. Search for the pattern [>@;]ee=float[;]. Expected error annotations are added by the synonymous options --fastq_eeout and --eeout described in vsearch-fastx_filter(1).

--xlength
Strip sequence length annotations from sequence headers when writing fasta or fastq files. Search for the pattern [>@;]length=integer[;]. Sequence length annotations are added by the --lengthout option.

--xsize
Strip abundance annotations from sequence headers when writing fasta or fastq files. Search for the pattern [>@;]size=integer[;]. Abundance annotations are added by the --sizeout option.

ignored options

--threads positive non-null integer
Command is not multithreaded, option has no effect.

EXAMPLES

Dereplicate sequences in input.fasta, annotate with abundance, write to derep.fasta, and record clustering details in derep.uc:

vsearch \
    --fastx_uniques input.fasta \
    --sizeout \
    --fastaout derep.fasta \
    --uc derep.uc

Dereplicate a fastq file, keeping only the best quality score at each position:

vsearch \
    --fastx_uniques input.fastq \
    --fastq_qout_max \
    --sizeout \
    --fastqout derep.fastq

SEE ALSO

vsearch-derep_fulllength(1), vsearch-derep_id(1), vsearch-derep_prefix(1), vsearch-derep_smallmem(1), vsearch-rereplicate(1), vsearch-fasta(5), vsearch-fastq(5), vsearch-expected_error(7)

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.

COPYRIGHT

Copyright (C) 2014-2026, Torbjørn Rognes, Frédéric Mahé and Tomás Flouri

All rights reserved.

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:

  1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

  2. 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: