RHEA reactions extraction

Dependecies:

• R (> 3.5)
• tidyverse (> 1.2.1)
• curl (> 7.xx)

Clone the git repository

mkdir /path/to/repo
cd /path/to/repo
git clone https://github.com/metaganal/rhea_rmd.git .

Data are already included in the data directory, but you can update to the latest RHEA DB through the following procedure.

Obtain RHEA DB reactions from RHEA SPARQL end-point

curl -H 'Accept: text/tab-separated-values' --data-urlencode 'query@rhea_sparql_query' https://sparql.rhea-db.org/sparql > data/rhea_db.tsv

Wrangle RHEA DB with R to get a table that's easier to process.

Rscript R/rhea_wrangling.R data/rhea_db.tsv

Auto curation of RHEA based on frequency of compound usage. Compounds with high frequency usage are removed as generic cofactors. Some compounds are used to define the directions of enzymatic reaction.

Rscript R/rhea_table_curation.R data/rhea_db_reactions.tsv data/rhea_db_parsed.tsv

List of outputs:

• data/rhea_db_parsed.tsv : RHEA DB table (parsed from SPARQL output)
• data/rhea_db_reactions.tsv : Reaction pairs that include cofactors and generic compounds
• data/rhea_reactions_* : Curated reaction pairs (only main substrate and product, annotated with reaction direction)
• data/rhea_reactants_* : Pairing of main substrate and products
• data/rhea_cofactor_* : Pairing of cofactors in the reaction (cofactors decided based on compounds usage frequency in the DB)
• data/rhea_generic_* : Reaction pairs with commonly used compound as substrate and products
• data/rhea_compound_usage : Frequency of compound usage

MOL files for Morgan Fingerprint/RDKit input can be downloaded from RHEA DB