NGLess Benchmarking

This repository includes code for benchmarking ngless.

This is a companion repository to:

NG-meta-profiler: fast processing of metagenomes using NGLess, a
domain-specific language by Luis Pedro Coelho, Renato Alves, Paulo Monteiro,
Jaime Huerta-Cepas, Ana Teresa Freitas, and Peer Bork in Microbiome 7:84 (2019)
http://doi.org/10.1186/s40168-019-0684-8

Reproducing

From scratch

1. Download all data: run the script download-data.py. This will download
   from ENA into a directory data/ (this process can take several hours and will
   result in ca. 60GiB of data).

Using the precomputed results

1. Precomputed data is available in data/precomputed
2. The script sources/plot-comparison.py generates the comparison plot