In teamcgc/cgcR: NCI Cancer Genomics Cloud examples in R

This tutorial provides the steps for creating a WiDdLe (WDL) file to run analysis on the FireCloud. A basic introduction to the FireCloud can be found here - https://github.com/teamcgc/cgcR/blob/master/vignettes/Firecloud.intro.Rmd

This example is similar to another tutorial created for the Seven Bridges Genomic (SB-CGC) platform - https://github.com/teamcgc/cgcR/blob/master/vignettes/UsingDocker.Part1.Rmd

What is WDL?

WiDdLe (WDL) is a Workflow Description Language developed by the Data Science & Data Engineering group at Broad Institute. It is used to describe tasks and workflow on the FireCloud. This tutorial will demonstrate how to
1) install everything you need to start writing and running WDL.
2) generate a JSON from WDL.
3) executing a WDL script, focusing on using Broad Institute's execution engine, which is called Cromwell.

References
https://software.broadinstitute.org/wdl/userguide/index
https://github.com/broadinstitute/wdltool

Installing everything you need to write and run WDL

Install JAVA, cromwell, sbt, and wdltool

brew install Caskroom/cask/java
brew install cromwell
brew install sbt

git clone https://github.com/broadinstitute/cromwell
cd cromwell #cd into git-cloned directory
sbt assembly
# `sbt assembly` will build a runnable JAR in `target/scala-2.11/cromwell-0.20.jar`

git clone https://github.com/broadinstitute/wdltool
sbt assembly
# `sbt assembly` will build a runnable JAR in `target/scala-2.11/wdltool-0.5.jar`

Start Docker - we will be using the latest Ubuntu image

Please refer to https://github.com/teamcgc/cgcR/blob/master/vignettes/UsingDocker.Part1.Rmd for using Docker

docker-machine start dev
docker-machine regenerate-certs dev
docker-machine env dev
eval $(docker-machine env dev)

Creating a WDL script

This WDL script use the command "grep" to find a string pattern in an input file.

Copy-and-paste the text below to a file named "grep.test.wdl"

task grep {

  String pattern
  File in_file

  command {
    grep '${pattern}' ${in_file}
  }

  runtime {
    docker : "ubuntu:latest"
  }

  output {
    File response = stdout()
  }
}

workflow test {
  call grep
}

Use wdltool to create json file that defines the search pattern ("KX") and input file ("temp.fasta").

The content of the input file ("temp.fasta") is

>gb:KU991811|Organism:Zika virus Brazil/2016/INMI1-Asian|Segment:null|Subtype:Asian|Host:Human
MKNPKKKSGGFRIVNMLKRGVARVSPFGGLKRLPAGLLLGHGPI

>gb:KU955590|Organism:Zika virus Z16019-Asian|Segment:null|Subtype:Asian|Host:Human
MKNPKKKSGGFRIVNMLKRGVARVSPFGGLKRLPAGLLLGHGPI

>gb:KU922960|Organism:Zika virus MEX/InDRE/Sm/2016-Asian|Segment:null|Subtype:Asian|Host:Human
MKNPKKKSGGFRIVNMLKRGVARVSPFGGLKRLPAGLLLGHGPI

>gb:KX056898|Organism:Zika virus Zika virus/GZ02/2016-Asian|Segment:null|Subtype:Asian|Host:Human
MKNPKKKSGGFRIVNMLKRGVARVSPFGGLKRLPAGLLLGHGPI

>gb:KU922923|Organism:Zika virus MEX/InDRE/Lm/2016-Asian|Segment:null|Subtype:Asian|Host:Human
MKNPKKKSGGFRIVNMLKRGVARVSPFGGLKRLPAGLLLGHGPI

>gb:KU963574|Organism:Zika virus ZIKV/Homo sapiens/NGA/IbH-30656_SM21V1-V3/1968-West_African|Segment:null|Subtype:West_African|Host:Human
MKNPKKKSGGFRIVNMLKRGVARVNPLGGLKRLPAGLLLGHGPI

In this example, the tool searches input file for a string "KX" and outputs the line containing the string.

Create a json file to define the search pattern and input file

java -jar wdltool/target/scala-2.11/wdltool-0.5.jar inputs grep.test.wdl > grep.test.json

Modify the json file so it looks like -

{
  "test.grep.pattern": "KX",
  "test.grep.in_file": "temp.fasta"
}

Run the WDL script

java -jar cromwell/target/scala-2.11/cromwell-0.20.jar run grep.test.wdl grep.test.json

Last few lines of the run should look like this - ```{} { "outputs": { "test.grep.response": "cromwell-executions/test/13f69bf7-449b-4e24-a324-bdf579c7ef07/call-grep/stdout" }, "id": "13f69bf7-449b-4e24-a324-bdf579c7ef07" } [2016-07-05 19:10:24,57] [info] SingleWorkflowRunnerActor workflow finished with status 'Succeeded'.

An output file is created at this location "cromwell-executions/test/13f69bf7-449b-4e24-a324-bdf579c7ef07/call-grep/stdout"  
The content of the output file is
```{}
>gb:KX056898|Organism:Zika virus Zika virus/GZ02/2016-Asian|Segment:null|Subtype:Asian|Host:Human

teamcgc/cgcR documentation built on May 31, 2019, 7:56 a.m.