The following lines demonstrates how to setup Sequence conservation on local computer using Docker.

Clone this repository:

git clone https://github.com/skodapetr/sequence-conservation.git

Enter into the new folder and run Docker build:

cd sequence-conservation
docker build \
    -t conservation \
    --build-arg USER=$(id -u ${USER}) \
    .

Run Docker image with mapping data directory into /data/conservation on host:

docker run \
    -v /data/conservation:/data/conservation \
    -u $(id -u ${USER}):$(id -g ${USER}) \
    -it conservation

In the next step we need to prepare a database for BLAST. There are two options, the database can be created locally on downloaded. Use one of the following command to prepare the database:

./download_database.sh

or

./prepare_database.sh

We can showcase use of the script by calculating conservation for 2SRC. First we enter the /data/conservation directory and download the FASTA file from PDB, next we run the Python script that computes the conservation.

cd /data/conservation
curl "https://www.rcsb.org/pdb/download/downloadFastaFiles.do?structureIdList=2SRC&compressionType=uncompressed" -o 2SRC.fasta
python3 /opt/conservation/calculate_conservation.py --input ./2SRC.fasta --output ./2SRC.json

You can close the Docker container by typing exit. When the Docker container is terminated all data outside volumes are lost. As we mapped the /data/conservation directory, the data in this directory are preserved and available from the host computer.

Another advantage of mapping of /data/conservation is that the BLAST database is preserved and does not need to be rebuild each time a new Docker container is started.

The initial content of this repository was heavily inspired by calc_protein_conservation.