AnnotationWebsite

Website on genome annotation

Configuration

Dependencies

This website is built using Django

• Poetry for Python dependencies management.
• PostgreSQL for database management.

Django

To use django the following environment variables should be properly set and exported:

export DJANGO_SECRET_KEY="a django secret key"
export PG_DBNAME="the database name"
export PG_USER="the postgres user who owns the database"
export PG_PASSWORD="PG_USER'S password"

To run the Unit tests an additional environment variable should be set:

export GITHUB_WORKSPACE="/the/path/to/the/repo's/root/on/your/machine"

You can add the previous variable exports to your ~/.bashrc, so that the variables are automatically loaded each time you open a new terminal.

PostgreSQL

sudo -u postgres psql --command="CREATE USER $PG_USER;"
sudo -u postgres psql --command="ALTER USER $PG_USER WITH ENCRYPTED PASSWORD '$PG_PASSWORD';"
sudo -u postgres psql --command="ALTER ROLE $PG_USER SET client_encoding TO 'utf8';"
sudo -u postgres psql --command="ALTER ROLE $PG_USER SET default_transaction_isolation TO 'read committed';"
sudo -u postgres psql --command="ALTER ROLE $PG_USER SET timezone TO 'UTC';"
sudo -u postgres psql --command="ALTER ROLE $PG_USER WITH CREATEDB;"
sudo -u postgres createdb --owner="$PG_USER" "$PG_DBNAME"
PGPASSWORD="$PG_PASSWORD" psql --username="$PG_USER" --host=localhost --list

Usage

Running the server and performing administrative tasks such as importing new genomes, go into the project's BASE_DIR. This is the directory created when executing django-admin startproject {{projectname}}. In our case, that is the Python/prokaryote directory.

cd Python/prokaryote

There you will find an executable file named manage.py. That is django's swiss-army knife. All commands related to testing, debugging, data import, and database management should be run via manage.py.

manage.y can be run in either of the following ways (the -h option displays a list of available subcommands):

./manage.py -h      # way 1: direct invocation
python manage.py -h # way 2: via Python (poetry's virtual env should be activated)
poetry run python manage.py -h # way 3: no need to activate the venv

Run tests (unit and integration)

# CAVEATS: This is still in development, do not execute
# --exclude-tag changes whilst in development
./manage.py test -v 2 --no-input --reverse # --exclude-tag strict 

Create the database (via dbexec)

To execute sql scripts use the command dbexec For example to create all the necessary tables, use :

python manage.py dbexec $GITHUB_WORKSPACE/Database/create-schema.sql  

Import data

In order to annotate genomes, we should have some genomes available, right? FASTA files can easily be imported via the command line.

Genomes (via importgenome)

ATTENTION: This subcommand is configured to import one genome at a time. If your file contains more than one FASTA entry (i.e. lines starting with >), the command will fail with an informative error message.

./manage.py importgenome $GITHUB_WORKSPACE/Data/Escherichia_coli_str_k_12_substr_mg1655.fa --specie "Escherichia coli" --strain k12
./manage.py importgenome $GITHUB_WORKSPACE/Data/Escherichia_coli_o157_h7_str_edl933.fa --specie "Escherichia coli" --strain edl933
./manage.py importgenome $GITHUB_WORKSPACE/Data/Escherichia_coli_cft073.fa --specie "Escherichia coli" --strain cft073
./manage.py importgenome $GITHUB_WORKSPACE/Data/new_coli.fa

Genes (with their annotation) (via importgenomes)

./manage.py importgenes $GITHUB_WORKSPACE/Data/Escherichia_coli_str_k_12_substr_mg1655_cds.fa
./manage.py importgenes $GITHUB_WORKSPACE/Data/Escherichia_coli_o157_h7_str_edl933_cds.fa
./manage.py importgenes $GITHUB_WORKSPACE/Data/Escherichia_coli_cft073_cds.fa
./manage.py importgenes $GITHUB_WORKSPACE/Data/new_coli_cds.fa

Proteins (via importproteins)

./manage.py importproteins $GITHUB_WORKSPACE/Data/Escherichia_coli_str_k_12_substr_mg1655_pep.fa 
./manage.py importproteins $GITHUB_WORKSPACE/Data/Escherichia_coli_o157_h7_str_edl933_pep.fa
./manage.py importproteins $GITHUB_WORKSPACE/Data/Escherichia_coli_cft073_pep.fa
./manage.py importproteins $GITHUB_WORKSPACE/Data/new_coli_pep.fa