FastKit

Fastkit is primarily written to perform routine/repetitive operations on FASTA and FASTQ files, most notably formatting. Its purpose is pre-processing of files for use in other bioinformatics tools.

• FastKit seeks to wrap established libraries (such as Biopython, SeqTK, FASTX-toolkit, etc) rather than re-invent the wheel!
• Write output to stdout so that the consumer can control file names transparently and pipe between fastkit subcommands to the target tool/file.
• Each subcommand should infer datatypes from filenames and process accordingly, rather than having a tool for each datatype.
• Raise error messages that provide good user experience

---

Example usage

fastkit format input.raw.fasta --strip-header-space > input.fasta
fastkit validate input.raw.fasta --dna  # Will raise an error if not valid IUPAC DNA

Testing in development

# Does not require pip install of fastkit
python fastkit/format.py --strip-header-space test/data/spaces.fas

Running tests

python -m unittest tests/*.py

# Or

./run_tests.sh

---

Available datatypes

• FASTA

Available subcommands

• format
• validate

Format

usage: fastkit format [-h] [--strip-header-space] [--uppercase] filename

Reformat FASTA files in preparation for tool execution.

Available filters:
- Strip spaces from FASTA headers
- Convert sequence characters to uppercase

# TODO: filter escaped chars from Galaxy text input

positional arguments:
  filename              A filename to parse and correct.

options:
  -h, --help            show this help message and exit
  --strip-header-space  Strip spaces from title and replace with underscore
  --uppercase           Transform all sequence characters to uppercase

Validate

usage: fastkit validate [-h] [--protein] [--dna] [--no-unknown] [--sequence-count SEQUENCE_COUNT] filename

Validate FASTA files in preparation for tool execution.

These functions should not alter contents but only raise exceptions or return
boolean values to communicate validity of data.

Available validators:
- dna
- protein
- no-unknown
- sequence-count

positional arguments:
  filename              A filename to parse and correct.

options:
  -h, --help            show this help message and exit
  --protein             Validate as IUPAC protein sequence
  --dna                 Validate as IUPAC DNA sequence
  --no-unknown          Prohibit unknown IUPAC characters (X/N) - requires --dna or --protein
  --sequence-count SEQUENCE_COUNT
                        [int] Maximum number of sequences that are permitted

---

Adding a subcommand

• Create new function(s) in fastkit/<new_command>.py
• <new_command>.py must have a main callable - use format.py as an example
• Import new_command in fastkit.__init__.py
• Add new_command to fastkit.cli.SUBCOMMANDS

Pushing changes to bioconda

N.B. bioconda-bot will make automated pull requests to update fastkit when a new release has been published on GitHub

• Publish a new release on GitHub
• Fork bioconda/bioconda-recipes and make a branch for the new version
• Update the version and sha256 in recipes/fastkit/meta.yaml to match the new release
• Commit, push and make a pull request to bioconda/bioconda-recipes
• Wait for it to be merged (you may need to ask bioconda-bot to add a label once it’s been approved)