README.md

• Install environment: conda env create --file environment.yaml
• Activate environment: conda activate snakemake
• Run analysis:

cd workflows/{analysis of choice}
snakemake --forceall --cores

Results

csv-tables containing the similarities between all samples are in the workflows/{comparison}/aggregated_sim folder.

Conducted comparisons are:

• megago based on real-world data
• string based comparison on real-world data
• megago based on random sets of go terms of various sizes

real world results

Input data is from here

using megago

• biological process plot
• cellular component plot

using string based comparison

according to methods in preprint of metaproteomics tool survey

• biological process plot
• cellular component plot

random set comparison

GO-terms are sampled from SwissProt to resemble real-world data. Two sets of equal size are drawn n times, repeated for different sample sizes.