Formatting & content WIP
[rnaseq-cwl-training.git] / setup.md
index b8c50321d8b07f8a76f8e925416957c3f274012e..f907ec716ffe9f6e5bbf86725bf02b1ee7d8e631 100644 (file)
--- a/setup.md
+++ b/setup.md
@@ -1,7 +1,75 @@
 ---
 title: Setup
 ---
-FIXME
+
+# Setting up a practice repository
+
+We will create a new git repository and import a library of existing
+tool definitions that will help us build our workflow.
+
+Create a new git repository to hold our workflow with this command:
+
+```
+git init rnaseq-cwl-training-exercises
+```
+
+On Arvados use this:
+
+```
+git clone https://github.com/arvados/arvados-vscode-cwl-template.git rnaseq-cwl-training-exercises
+```
+
+Next, import bio-cwl-tools with this command:
+
+```
+git submodule add https://github.com/common-workflow-library/bio-cwl-tools.git
+```
+
+# Downloading sample and reference data
+
+Start from your rnaseq-cwl-exercises directory.
+
+```
+mkdir rnaseq
+cd rnaseq
+wget --mirror --no-parent --no-host --cut-dirs=1 https://download.pirca.arvadosapi.com/c=9178fe1b80a08a422dbe02adfd439764+925/
+```
+
+# Downloading or generating STAR index
+
+Running STAR requires index files generated from the reference.
+
+This is a rather large download (4 GB).  Depending on your bandwidth, it may be faster to generate it yourself.
+
+## Downloading
+
+```
+mkdir hg19-chr1-STAR-index
+cd hg19-chr1-STAR-index
+wget --mirror --no-parent --no-host --cut-dirs=1 https://download.pirca.arvadosapi.com/c=02a12ce9e2707610991bd29d38796b57+2912/
+```
+
+## Generating
+
+Create `chr1-star-index.yaml`:
+
+```
+InputFiles:
+  - class: File
+    location: rnaseq/reference_data/chr1.fa
+    format: http://edamontology.org/format_1930
+IndexName: 'hg19-chr1-STAR-index'
+Gtf:
+  class: File
+  location: rnaseq/reference_data/chr1-hg19_genes.gtf
+Overhang: 99
+```
+
+Generate the index with your local cwl-runner.
+
+```
+cwl-runner bio-cwl-tools/STAR/STAR-Index.cwl chr1-star-index.yaml
+```
 
 
 {% include links.md %}