Requirements

Install the Google Cloud SDK on your system.

https://cloud.google.com/sdk/docs/install

Configure access to BigQuery

In the google cloud console, under IAM & Admin -> IAM, grant your user the "BigQuery User" role.

Create a service account and key

In the google cloud console, under IAM & Admin' -> 'Service Accounts, create a service account. Then select that service account and choose Add Key. This will download a json file with your private key information. Make a copy of this file in the server subdirectory and name it credentials.json

Ingest sequences

I am using Google BigQuery to store the sequences. In the bq subdirectory is a schema file sequence_schema.json that defines the BQ schema and a script that will load the data loaddata.sh. The script assumes that all the individual sequence json files are in a local directory. This has scalability limitations. For a scalable solution, the files should be loaded from google cloud storage.

• put the json sequence files on your local system and use this location as the <path-to-files> parameter below

• from the terminal, log into gcloud and follow the instructions to authenticate gcloud auth login

• set your project gcloud config set project <project-id>

• from the bq directory, run the loaddata shell script. You can you any text names (with _ as the only special character allowed) for the dataset and table parameters. The script will create them if they do not exist. Specify a filename as which will be used to correctly format the data for upload. This load command is not idempotent. It will load the data as many times as it is called with the same dataset and table parameters.

  ./loadddata.sh <google-project> <dataset> <table> <path-to-files> <tmpfile>

Run the client and server

From the main directory run:

docker-compose build

docker-compose up

At any time to stop the client and server, you can run:

docker-compose stop

or you can kill the docker and then run:

docker-compose down

Use the Sequence-Finder UI

On the system running the docker image, go to: http://localhost:3031

Click the Edit button to specify the project, dataset and table where the sequences are loaded.

Input any combination of nucleotides (A, C, T, G, a, c, t, g) and click on submit (currently pressing the Enter key while in the text box does not work). The field will only allow these characters.

REST API

The REST API is directly accessible at http://localhost:3003

Supported API:

HTTP Method	Path	Functionality
GET	/dbinfo	Retrieves the google project, dataset and table containing the seqeunce data
POST	/dbinfo/project/dataset/table	Sets the google project, dataset and table to use to query sequences
GET	/sequence/id	Retrieves a sequence by its id
GET	sequence/match/bases	Returns all of the sequence objects where the sequence contains the bases

Run server tests

From the server subdirectory, run npm run test