Annotator for enzymes using an ensemble of tools to annoatte function to genes.
conda create --name annotatee python=3.11
pip install annotatee
pip install enzymetk
conda install -c bioconda -c conda-forge diamond
conda install -c conda-forge -c bioconda foldseek
Download sequences for your database (e.g. swissprot as a fasta file) e.g. the fasta file from UniProt SwissProt with annotations. Format:
>someid1
MRWQEMGYIF
>someid2
MRWQEMGYIF
Pass a fasta file as the database and a fasta file that you seek to search against the database.
The database will first get searched for existing anontations, and otherwise ML methods will be used.
Example command:
annotatee fasta query.fasta ref.fasta --methods blast,foldseek --run-name example_output
(see example fasta files in the example_data folder, run the above command with them!)
annotatee fasta input_df.csv Uniprot_reviewed_catalytic_activity_06032025.fasta --methods blast,foldseek --foldseek-db foldseek/structures/pdb/pdb --output-folder output/ --run-name omgprot50
Note you may have issues with dependencies e.g. proteiner requires python 3.7 which needs special channels on mac... these ones don't work as easily...
git clone git@github.com:ArianeMora/annotate-e.git
source install_scripts/clean.sh
source install_scripts/proteinfer.sh
You'll need to pass the directory of the install for CLEAN and proteInfer to annotatee. e.g. --clean-dir ...../CLEAN/app/
Note! requires enzymetk to also be installed: enzymetk.
pip install enzymetk
For now to run with CLEAN and proteInfer please follow their instructions respectfully (I will make this simpler when I can do a data drop on Zenodo - if you want to use our versions, please email or post an issue).
Install Proteiner and Install CLEAN
annotattee --help
Usage: annotatee [OPTIONS] QUERY_FASTA DATABASE
Find similar proteins based on sequence or structural identity in order to annotate these using BLAST and FoldSeek. Also annotate with ProteInfer and CLEAN.
╭─ Arguments ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ * query_fasta TEXT Full path to query fasta or csv (note have simple IDs otherwise we'll remove all funky characters.) [default: None] [required] │
│ * database TEXT Full path to database fasta (for BLAST and FoldSeek) [default: None] [required] │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭─ Options ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ --output-folder TEXT Where to store results (full path!) [default: Current Directory] │
│ --run-name TEXT Name of the run [default: annotatee] │
│ --clean-dir TEXT Directory (full path) to CLEAN - if not using the default) │
│ --proteinfer-dir TEXT Directory (full path) to proteiner - if not using the default) │
│ --run-method TEXT Run method (filter or complete) i.e. filter = only annotates with the next tool those that couldn't be found. [default: complete] │
│ --keep-dups --no-keep-dups Whether or not to keep multiple predicted values if False only the top result is retained. [default: no-keep-dups] │
│ --args-blast TEXT comma separated list (no spaces) of arguments to pass to Diamond BLAST │
│ --args-foldseek TEXT comma separated list (no spaces) of arguments to pass to foldseek │
│ --args-proteinfer TEXT comma separated list (no spaces) of arguments to pass to ProteInfer │
│ --args-clean TEXT comma separated list (no spaces) of arguments to pass to CLEAN │
│ --methods TEXT comma separated list (no spaces) of methods to run (e.g. could just pass ['foldseek', 'proteinfer']) to pass to CLEAN │
│ --foldseek-db TEXT Database for foldseek to override fasta before (e.g. path to all pdbs as per foldseek docs.) │
│ --id-col TEXT id column in df if df passed (csv) rather than fasta [default: id] │
│ --seq-col TEXT Database for foldseek to override fasta before (e.g. path to all pdbs as per foldseek docs.) [default: seq] │
│ --install-completion Install completion for the current shell. │
│ --show-completion Show completion for the current shell, to copy it or customize the installation. │
│ --help Show this message and exit. │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯