Overview

SeqFu consists of core programs, accessible as subcommands, and a set of corollary utilities, having the prefix fu-.

Type seqfu alone to list the core subprograms, and seqfu {command} --help to access the help of each specific subcommand.

To simply print the version type seqfu version and to print the citation seqfu cite.

Basic operations: cat, head, tail, grep, rc

These commands have been inspired by the common GNU commands, and all can read from the standard input.

Their usage is quite intuitive, so here we highlight some special feature.

seqfu cat can read mixed FASTA and FASTQ files, and be forced to output in either FASTA or FASTQ format. Some basic manipulations are bundled, like:

  • Forcing FASTA or FASTQ output
  • Manipulating sequence name (prefix, suffix, prepend filename, remove comments…)
  • Add infos in the header (length, gc content, original name)
  • Filter by length (minimum length, maximum length, trim bases at the beginning or the end…)
  • …

seqfu grep can be used to extract sequences by matching oligonucleotides, that would be scanned also in the reverse strand and allowing for mismatches or partial matches. The oligo can be in IUPAC alphabet with ambiguous bases (e.g. degenerate primers).

seqfu head can skip a number of sequences (i. e. print the first N sequences taking one every M), to extract a small subset sampling deeper.

The reverse complement function (seqfu rc) is unique in taking as input both files and sequences, and properly supports IUPAC degenerate bases.

Getting an idea: view, qual, stats, count

seqfu view is only for interactive use, and can be used to have a visual feedback on the quality values and on the presence of oligonucleotides:

View

seqfu stats can print the total number of sequences, bases, average, N50, N75, N90 and AuN, minimum and maximum length of a dataset, both in TSV format and with a nicer console oriented output:

┌────────────────────┬───────┬──────────┬───────┬─────┬─────┬─────┬────────┬─────┬─────┐
│ File               │ #Seq  │ Total bp │ Avg   │ N50 │ N75 │ N90 │ auN    │ Min │ Max │
├────────────────────┼───────┼──────────┼───────┼─────┼─────┼─────┼────────┼─────┼─────┤
│ filt.fa.gz         │ 78730 │ 24299931 │ 308.6 │ 316 │ 316 │ 220 │ 0.385  │ 180 │ 485 │
│ illumina_1.fq.gz   │ 7     │ 630      │ 90.0  │ 90  │ 90  │ 90  │ 12.857 │ 90  │ 90  │
│ illumina_2.fq.gz   │ 7     │ 630      │ 90.0  │ 90  │ 90  │ 90  │ 12.857 │ 90  │ 90  │
│ illumina_nocomm.fq │ 7     │ 630      │ 90.0  │ 90  │ 90  │ 90  │ 12.857 │ 90  │ 90  │
└────────────────────┴───────┴──────────┴───────┴─────┴─────┴─────┴────────┴─────┴─────┘

Managing datasets: interleave, deinterleave, lanes

Very common tasks when dealing with Illumina Paired-End sequences are interleaving and deinterleaving the datasets.

seqfu interleave and seqfu deinterleave can do that, with high speed and lower corruption risks.

Multiple lanes can be quickly merged with seqfu lanes.

Sorting, dereplicating

seqfu sort can sort sequences by length.

seqfu derep can be used to dereplicate datasets, printing the number of identical sequences. In particular, this information can be used also from the input dataset, allowing to dereplicating a set of dereplicated files keeping trace of the number of sequences.