Analysis

Scripts and global analysis of eggNOG-mapper performance compared against BLAST, InterProScan and CAFA2. Genomes and metagenomes evaluated.

Supplementary Material for the paper: "Fast genome-wide functional annotation through orthology assignment by eggNOG-mapper" http://biorxiv.org/content/early/2016/09/22/076331

Analysis

Additional figures and details are available as jupyter notebook in this reposotory.

Reproducibility

Benchmark required running BLAST, InterProScan and eggNOG-mapper independently and, in the case of eggNOG-mapper, using multiple combinations of parameters.

All raw data (~55GB uncompressed) can be downloaded from: http://beta-eggnogdb.embl.de/download/eggnog_4.5/eggnog-mapper-data/emapper_benchmark_data.tar.gz

Data includes 4 self-explanatory directories:

  blast/
  interpro/
  emapper_hmm/
  emapper_diamond/

To re-run the whole benchmark analysis, decompress all 4 directories under within this repository, and execute benchmark.py, which will generate a digested all_benchmark_tables.pkl file to be used within the jupyter notebook.

Name		Name	Last commit message	Last commit date
Latest commit History 32 Commits
CAFA2		CAFA2
data		data
metagenomes		metagenomes
plots		plots
README.md		README.md
all_benchmark_tables.pkl		all_benchmark_tables.pkl
augment_go_terms.py		augment_go_terms.py
benchmark.py		benchmark.py
benchmark_analysis.ipynb		benchmark_analysis.ipynb
benchmark_summary.html		benchmark_summary.html
blast_annotate.py		blast_annotate.py
dump_groundtruth.py		dump_groundtruth.py
go.tree		go.tree
plot_functions.py		plot_functions.py
run_blast.sh		run_blast.sh
run_emapper.sh		run_emapper.sh
supplementary_material.ipynb		supplementary_material.ipynb

eggnogdb/emapper-benchmark

Folders and files

Latest commit

History

Repository files navigation

Analysis

Reproducibility

About

Resources

Stars

Watchers

Forks

Languages