Reference k-mer distribution files for Bracken
======

Bracken re-estimates abundance of taxonomic units at given rank based on
Kraken's classification result.

 - Website: https://ccb.jhu.edu/software/bracken/
 - Citation: Lu J, Breitwieser FP, Thielen P, Salzberg SL. Bracken: estimating
   species abundance in metagenomics data. PeerJ Computer Science. 2017.
   3:e104.

Built using Bracken v2.7. Command:

```
for l in 50 75 100 150 250 300; do
  bracken-build -d . -t # -k 35 -l $l
done
```

Notes:

 - This database is based on the corresponding Kraken2 database. See:
   ../kraken2/.

 - l = 50..300 are the target read lengths. One should select an appropriate
   distribution file according to the input data.

 - k = 35 (bp) is the default k-mer length for Kraken 2, while that for
   Kraken 1 is 31.

 - database.kraken is an intermediate file to facilitate creation of custom
   distribution files.

Typical re-estimation command (assuming read length is 150 bp and target rank
is species (S):

```
bracken -d . -i input.kreport -o output.bracken -r 150 -l S
```