MinEncCanKmer

Minimal encoding of canonical k-mers

Compilation

gcc -O3 -D_FILE_OFFSET_BITS=64 -pthread -mbmi -o canonical fasta.c canonical.c -lm

Usage

./canonical <fasta> <k> <b>

where

• fasta is a (multiple) fasta file
• k is the k-mer length, optional, default 5, maximum 31
• b is the number of buckets, optional, default 4

Output: distribution of k-mers to buckets

2-bit encoding

To switch to standard 2-bit encoding, (un)comment the following lines:

// process_string(seq,k,threads,t)
   process_string_std(seq,k,threads,t)

General alphabets

For encoding canonical k-mers on general (non-DNA) alphabets, Python scripts are provided in the according subfolder, where minenc.py outputs the encoding of all k-mers for a given alphabet size, and minenc_rc.py encodes considering reverse complementation.

C++ implementation

An implementation of the functionality in C++ is available in the genesis library, see here. The library also offers other useful functionality for working with sequences and k-mers.

Citation

Please cite: Wittler R. General encoding of canonical k-mers. Peer Community Journal. 2023;3: e87. https://doi.org/10.24072/pcjournal.323

License

• fasta.c and fasta.h are borrowed from FragGeneScan-Plus.
• MinEncCanKmer is licensed under the GNU general public license.