Skip to content

Latest commit

 

History

History
52 lines (35 loc) · 1.54 KB

README.md

File metadata and controls

52 lines (35 loc) · 1.54 KB

fastalen

Every bioinformatician has its own method to get the size of sequences from FASTA files. Here is mine.

Compilation

fastalen is written in C99, and uses the POSIX C library. In addition, it uses:

  • kseq.h from klib to parse FASTA sequences,
  • argtable3 to parse command-line arguments.

This files are included.

To compile, just run

$ c99 -o fastalen argtable3.c fastalen.c

Usage

Usage: fastalen  <in.fasta> [<in.fasta>]...  [-t|--tsv] [-c|--csv] [-C|--count] [-O|--only-count]  [--help] [--version]
Read file(s) containing FASTA sequence(s) and return their ID and length.

Mandatory argument:         
  <in.fasta>                input fasta file(s)
                        
Formatting options:
  -t, --tsv                 output as TSV
  -c, --csv                 output as CSV
  -C, --count               count the number of sequences
  -O, --only-count          ONLY count the number of sequences
  -s, --sum                 sum the size of the sequences
                        
  --help                    display this help and exit
  --version                 display version info and exit

Licence

Copyright © 2018 Sylvain PULICANI [email protected]

This work is free. You can redistribute it and/or modify it under the terms of the Do What The Fuck You Want To Public License, Version 2, as published by Sam Hocevar. See the COPYING file for more details.