Dummy binner of contigs and reads that uses kmer frequencies, GC content and coverage. Few dependencies, basic results.

Dummy binner is GNU publicly licensed. Enjoy it.

In case you cite it:

Santos-Junior, C.D. (2023) Dummy binner - Using tetranucleotide frequencies, GC and
coverage to cluster contigs and reads. Software in GitHub Repository available
in <https://github.com/celiosantosjr/dummybinner>. Accessed in Month/Year.

Dummy binner uses python>3.9 and a set of packages easilly installed using

$ pip install biopython numpy pandas scipy scikit-learn tqdm pysam

The versions required follow:

usage: dummy_binner2.py [-h] [--r R] --mode MODE [--otag OTAG]
                        [--minlen MINLEN] [--k K] [--threshold THRESHOLD]
                        [--mincontigs MINCONTIGS] [--cov COV]
                        (--infile INFILE | --f F)

Dummy binner can be used in the reads mode, where it accepts the minimum sequence length (--minlen), the R1 and R2 files (--f and --r, respectively), and you can also specify the output tag for the output filename (--otag). The basic usage cases are:

python3 dummy_binner2.py --f <R1.fq.gz> --r <R2.fq.gz> --mode reads

For contigs, the binning is a bit more parameterized, you should input a contigs file (--infile), can also specify the minimum contig length to be accepted (--minlen) and the minimum number of contigs in a bin to be considered (--mincontigs). You also can alternatively give the kmer size (--k) and the maximum distance between two contigs (--threshold). The basic usage cases are:

python3 dummy_binner2.py --infile <contigs.fa> --mode contigs

To generate coverage profiles to be applied in the binner, please use the script make_coverages.py. This script works in a set of bam files pre-sorted and indexed. The basic usage is as follows:

python3 make_coverages.py [-h] -c CONTIGS -o OUTPUT bam_files [bam_files ...]

In case any issues, please contact me.