This repository contains two projects:

• cromwell-client - a client library that interacts with Cromwell workflow engine from Java or browser
• _cromwell-web _ - a simple UI for running Cromwell workflows in the browser

Cromwell-Web

Cromwell-web subproject is a simple UI for accessing cromwell REST API. To run it open sbt console and go to CromwellWebJVM subproject and type

reStart

It is also published as a Docker container. You can run it as:

docker run -p 8001:8001 quay.io/comp-bio-aging/cromwell-web:0.5.2

Services

You can start cromwell, mysql and cromwell-client altogether by starting a docker stack:

cd services
stack deploy -c docker-compose.yml cromwell

Make sure that you have docker swarm initialized ( https://docs.docker.com/engine/reference/commandline/swarm_init/ ) on your machine before deploying the stack You can also use docker-compose instead of docker swarm.

cromwell-client

Scala/ScalaJS Client for the Cromwell workflow engine. This is a work-in-progress, not all methods have been implemented and no tests yet.

Adding to dependencies

add the following to you build.sbt

resolvers += "jitpack" at "https://jitpack.io"
libraryDependencies += "group.research.aging" %%% "cromwell-client" % "0.5.4"

Setting up files and host parameters

For setting up opening files/folders on the server you can to specify the FILE_PREFIX_URL environment variable. For example if the files can be accessed by http://cromwell-server then FILE_PREFIX_URL=http://file_server_url should be specified either as a parameter for docker container or docker-compose configuration. if not specified it shows /data folder. By default, at http://cromwell-client:port/data/<your_path> it exposes exposes /data folder of the file system. It can be changed by configuring DATA environment variable (which is /data by default)

Usage

Here are examples of the usage.

Get all metadataabout existing workflows:

import group.research.aging.cromwell.client._
val client = CromwellClient.localhost //or put with url of the cromwell server
val workflowsInfo: IO[List[Metadata]] = client.getAllMetadata()
//if you want to get as a Future
val workflowsInfoAsFuture: Future[List[Metadata]] = workflowsInfo.unsafeToFuture()
//if you want to get as a plain result (blocking)
val workflowsInfoBlocking: Seq[Metadata] = workflowsInfo.unsafeRunSync()

Note: in many methods IO monad is returned that can be easily turned into Future or plain (blocking) result.

Run workflow and get its status:

import java.io.{File => JFile}
import better.files._
import group.research.aging.cromwell.client._
import scala.concurrent.Future

val client = CromwellClient.localhost

val workflow = File("/data/sources/rna-seq/pipelines/quantification/quantification.wdl")
val inputs = File("/data/sources/rna-seq/pipelines/quantification/inputs/worms.json")
val result = client.postWorkflowFiles(workflow, inputs)

Get the outputs by the id:

import group.research.aging.cromwell.client._
val client = CromwellClient.localhost
val id = "548a191d-deaf-4ad8-9c9c-9083b6ecbff8"
val outputs = client.getOutputs(id)

CromwellWeb REST API

Cromwell Web also allows has some REST API to run bioinformatic workflows. It is similar to cromwell and call cromwell server underneath /api/ shows swagger interface with docs.

http://localhost:8001/api/all

http://localhost:8001/api/run/<workflows_name>/