MLSimian

Environment

1. We can run our application on the localhost or in the cloud, using a RDS instance, for example. For a production environment, change the analysis_config.yml and stats_config.yml, changing the database connection.

2. We can simulate our application on localhost with docker-compose. Run the command: docker-compose build. We should have 3 images:

• mysql_image (optional): You can use this image if you want a localhost connection
• analysis_service_image
• stats_service_image

3. We can run our applications with the command: docker-compose up. We should then have 3 running containers:

• mysql_image (optional): In a production environment, you could use a RDS instance, for example
• analysis_service_image: The service responsible for processing new Simian Analysis. (POST) => http://localhost:8080/simian
• stats_service_image: The service responsible for retrieving computed Simian Stats. (GET) => http://localhost:8090/stats

4. You also can redirect your machine localhost to your container localhost with the following nginx setup on sites-available folder:

server {
	listen 80 default_server;
	listen [::]:80 default_server;

	root /var/www/html;

	index index.html index.htm index.nginx-debian.html;

	server_name _;

	location / {
		proxy_pass  http://127.0.0.1:12345/;
	}
}

How to start the MLSimian applications (outside docker-compose)

1. Run mvn clean install to build your application
2. Create a mysql schema, for example ml_simian
3. Setup your database configurations (mysql user, mysql password, mysql host and schema name) inside the config.yml
4. Run database migrations with java -jar target/ml-simian-analysis-1.0.jar db migrate config.yml OR java -jar target/ml-simian-stats-1.0.jar db migrate config.yml
5. Start application with java -jar target/ml-simian-analysis-1.0.jar server config.yml OR java -jar target/ml-simian-stats-1.0.jar server config.yml
6. To check that your application is running enter url http://localhost:8080

Tests

There are unit tests in this project. You can run it with mvn clean test

Project structure

1. Resources are responsible for handling HTTP requests
2. Services are responsible for handling the business logic
3. DAOs are responsible for database connections

Endpoints

1. You can add a new DNA sequence with a POST method on the URL www.grcampos.com/simian: curl --header "Content-Type: application/json" --request POST --data '{"dna": ["ATGCGA", "CAGTGC", "TTATGT", "AGAAGG", "CCCCTB", "TCACTG"]}' http://www.grcampos.com/simian
2. You can check the results with a GET method on the URL www.grcampos.com/stats: curl --header "Content-Type: application/json" --request GET http://www.grcampos.com/stats

Other

1. The project is running on an AWS EC2 server and a AWS RDS. Both are very small machines and may fail. Please, feel free to contact me if the server is not reachable or if you need to clear the database.
2. For the coding algorithm, I assumed that the DNA sequence should be case insensitive. Besides, a sequence with more than 4 equal characters should still count as a single sequence
3. Invalid DNA sequences does not sum up on the count_human_dna property, but both bad formatted DNA's and human DNA's, returns a 403 status

Contact

grcampos21@gmail.com | (11) 99117-6997