Sequence utilities · Software
SeqKit
A cross-platform toolkit for fast inspection, filtering, conversion, and analysis of FASTA and FASTQ files.
What SeqKit is
SeqKit is an ultrafast, cross-platform command-line toolkit developed by Wei Shen, Botond Sipos, and Liuyang Zhao for processing and manipulating FASTA and FASTQ sequence data. Biological workflows regularly require filtering reads, calculating assembly metrics, searching sequence motifs, and reformatting large files, which can create bottlenecks in routine pipelines.
The software provides 39 subcommands covering basic sequence operations, format conversions, motif searches, set operations, sequence editing, and ordering. It reads and writes standard and compressed sequence streams seamlessly, serving as an everyday utility for bioinformatics and computational genomics.
How it works
SeqKit runs as a lightweight, statically linked binary requiring no external dependencies. It seamlessly parses both FASTA and FASTQ formats across standard input, standard output, and files compressed with gzip, xz, zstd, bzip2, or lz4. Gzip decompression and compression performance is accelerated using optimised libraries such as pgzip and compress.
Operations can be configured using regular expressions for custom sequence identification or replacement. Several commands offer strand sensitivity (processing the positive strand, both strands, or user-selected strands), and multithreading is utilised where applicable (defaulting to four threads). Stochastic commands such as sampling and shuffling accept user-defined random seeds to ensure reproducible outputs.
Key facts
- Supported inputs
- FASTA, FASTQ, TSV, BAM
- Supported compression formats
- gzip, xz, zstd, bzip2, lz4
- Subcommands
- 39 distinct commands
- Threading
- Default 4 threads; multi-threading supported on select subcommands
What you can use it for
- Sequence statistics and quality inspection
- Use the stats and watch commands to rapidly calculate metrics including sequence count, minimum and maximum lengths, N50 values, and Q20/Q30 quality base percentages.
- Sequence motif searching and amplicon extraction
- Search sequences by identifier or motif using grep and locate, or extract in silico PCR products using the amplicon command with mismatch allowances.
- Subsampling and paired-end read synchronisation
- Downsample large FASTQ datasets by read count or proportion using sample or sample2, and restore matched pairs across two paired-end read files using the pair command.
- Format conversion and quality score translation
- Convert between FASTA, FASTQ, and tab-delimited formats, or translate quality score encodings between Sanger, Solexa, and Illumina standards.
- Genomic sequence editing and circularisation
- Introduce point mutations, insertions, or deletions via mutate, concatenate matching sequence IDs, or reset the coordinate origin of circular genomes using restart.
Run it here
SeqKit is preinstalled on the machine behind every research conversation. Ask the agent in plain language, or give it the exact command — it runs it, reads the output and explains it.
Strengths and limits
Good at
- Processes uncompressed and compressed files (gzip, xz, zstd, bzip2, lz4) interchangeably through standard input and output pipelines.
- Provides 39 versatile subcommands addressing diverse tasks from indexing and searching to quality monitoring.
- Ensures reproducibility in randomised tasks such as sampling and sequence shuffling through configurable random seeds.
- Supports regular expressions for sequence identifier extraction, search, and replacement.
Watch out for
- Several ordering and file-splitting commands (such as sort, shuffle, and split) are designed primarily for FASTA and only prefer FASTA input.
- Select sequence transformation, replacement, and sliding commands operate only on the positive (forward) strand.
- Multi-threading acceleration is limited to specific commands such as stats, scat, sum, and partial search operations.
Questions
Can I run SeqKit online without installing it?
SeqKit is preinstalled on the machine behind every research conversation. Ask the agent in plain language, or give it the exact command — it runs it, reads the output and explains it.
What licence is SeqKit under?
SeqKit is published under its authors' own licence; the project page at https://bioinf.shenwei.me/seqkit/ states it.
Can SeqKit search for motifs on both the forward and reverse strands?
Yes. Search and pattern matching commands such as grep, locate, amplicon, and fish search across both the positive and negative strands and allow mismatches.
Which quality score encodings can SeqKit convert between?
The convert command transforms FASTQ quality score encodings across Sanger, Solexa, and Illumina formats.
Can SeqKit handle BAM files?
Yes. The bam subcommand enables monitoring and the generation of online histograms for BAM record features.
How are broken single-line FASTQ files fixed?
SeqKit includes the sana subcommand specifically to sanitize broken single-line FASTQ files.
How to cite it
Wei Shen*, Botond Sipos, and Liuyang Zhao. 2024. SeqKit2: A Swiss Army Knife for Sequence and Alignment Processing. iMeta e191. doi:10.1002/imt2.191.
Written from SeqKit's own README and project page. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.