A tool to store kmer unique to dataset

To install:

pip install kmeruniq 

Two commands:

• kmeruniq build
• kmeruniq query

Check the help with -h flag.

Usage example of the CLI

kmeruniq build -k 21 --fof my_file.fof --output path/to/index -e --shard 20 
kmeruniq query --query_str ACGAAACGTACATTCACACACACACATAGAGAAGGAGAGCAGCACACACA --index-path path/to/index
kmeruniq query --query_file path/to/some/data.fa --index-path path/to/index

The output is a the result of a Counter for each value.

The fof format

A file of file. That is a line separated list of file.

/path/to/foo.fa
/path/to/bar.fa
/path/to/barrr.fa.gz

Each line can also specify a label for the file. Two file with the same label will be considered as merge. For instance:

/path/to/foo.fa     ;chr1
/path/to/bar.fa     ;chr2
/path/to/barrr.fa.gz   ;chr1

Here the foo.fa and barr.fa.gz will be merged together within the index.

Remark that file can be gzip compressed.

Usage example of the Python API:

The Kmer and DNA datatype are the one used by vizibridge.

from kmeruniq.index import Index
from vizibridge import Kmer, DNA

idx = Index("path/to/my/index")
# idx is a dict-like object keyed by kmer and valued by the annotation

idx["ACG..ACG"] # some kmer of the appropriate size. 

for kmer in idx:
    print(idx[kmer]) # print the value of each kmer

dna = DNA("ACG....ACGT") # long sequence of DNA from somewhere

for kmer in dna.enum_canonical_kmer(idx.k):
    print(idx.get(kmer)) # print the value associated to kmer or None if kmer not inside.