Sequence: reformat FASTA line width

Operation

Rewrap FASTA records so each sequence is broken into fixed-width lines (the canonical 60/70/80-column convention) rather than one long line per record or whatever width an upstream tool emitted. The corpus uses one tool:

toolshed.g2.bx.psu.edu/repos/devteam/fasta_formatter/cshl_fasta_formatter (“FASTA Width formatter”).

The tool is part of the FASTX-Toolkit and can also fold case or emit single-line records, but the only mode the IWC corpus exercises is width rewrap — every occurrence is the mgnify amplicon family normalizing reads after quality control. Parameter names below are corpus-inferred from tool_state.

When to reach for it

• A downstream tool, viewer, or diff expects standard wrapped FASTA and an upstream step emitted single-line (or oddly wrapped) sequences.
• You are normalizing reads/contigs for distribution or for a reproducible test fixture and want stable line breaks.

If your need is editing headers rather than wrapping bodies, that is sequence-fasta-tabular-interconvert + relabel-fasta-headers-via-tabular, not this. Case folding and header rewriting are catalog capabilities of this tool with zero corpus uptake — don’t reach for them speculatively.

Parameters

• input: the FASTA (connected).
• width: characters per sequence line. Corpus value "60". 0 would emit each sequence on a single line (unused here).

Idiomatic shape

tool_id: toolshed.g2.bx.psu.edu/repos/devteam/fasta_formatter/cshl_fasta_formatter/1.0.1+galaxy2
tool_state:
  input: { __class__: ConnectedValue }
  width: "60"

Pitfalls

• Reformatting is not filtering or editing. This only changes line breaks; record count, ids, and sequence content are unchanged. If duplicates or short records are the problem, see sequence-merge-and-dedup or filter-by-length.
• Don’t use it as a FASTA validator. A malformed record may pass through reformatted; it is not a correctness gate.
• width: "0" collapses wrapping. Single-line-per-record output breaks tools that read fixed-width FASTA; only set it when a consumer explicitly wants unwrapped sequences.

See also

• sequence-merge-and-dedup — combine multiple FASTAs and drop duplicates.
• galaxy-sequence-patterns — the sequence pattern map.
• iwc-sequence-operations-survey — corpus evidence.