NAME
BioX::Workflow::Plugin::Drake - A very opinionated template based bioinformatics workflow writer for Drake.
SYNOPSIS
The main documentation for this module is at BioX::Workflow. This module extends Workflow in order to add functionality for outputing workflows in drake format.
biox-workflow.pl --workflow workflow.yml > workflow.drake
drake --workflow workflow.drake #with other functionality such as --jobs for asynchronous output, etc.
List your plugins in your workflow.yml file
---
plugins:
- Drake
global:
- indir: /home/user/gemini
- outdir: /home/user/gemini/gemini-wrapper
- file_rule: (.vcf)$|(.vcf.gz)$
- infile:
- min: 1 ##IF USING MIN
#So On and So Forth
More information about Drake can be found here https://github.com/Factual/drake.
Default Variables
BioX::Workflow::Plugin::Drake assumes your INPUT/OUTPUT and indir/outdirs are linked.
This means the output from step1 is the input for step2.
You can override this behavior by either declaring any of these values, or in the global variables set auto_input: 0, disable automatic indir/outdir naming with auto_name: 0, and disable automatically naming outdirectories by rule names with enforce_struct: 0.
Example
workflow.yml
---
plugins:
- Drake
global:
- indir: /home/user/workflow
- outdir: /home/user/workflow/output
- file_rule: (.csv)$
rules:
- backup:
local:
- INPUT: "{$self->indir}/{$sample}.csv"
- OUTPUT: "{$self->outdir}/{$sample}.csv"
- thing: "other thing"
process: |
cp $INPUT $OUTPUT
- grep_VARA:
local:
- OUTPUT: "{$self->outdir}/{$sample}.grep_VARA.csv"
process: |
echo "Working on {$self->{indir}}/{$sample.csv}"
grep -i "VARA" {$self->indir}/{$sample}.csv >> {$self->outdir}/{$sample}.grep_VARA.csv \
|| touch {$self->OUTPUT}
- grep_VARB:
local:
- OUTPUT: "{$self->outdir}/{$sample}.grep_VARA.grep_VARB.csv"
process: |
grep -i "VARB" {$self->indir}/{$sample}.grep_VARA.csv >> {$self->outdir}/{$sample}.grep_VARA.grep_VARB.csv || touch {$self->OUTPUT}
Notes on the drake.yml
Drake will stop everything if you're job returns with an exit code of anything besides 0. For this reason we have the last command have a command1 || command2 syntax, so that even if we don't grep any "VARB" from the file the workflow could continue.
Run it with default setup
biox-workflow.pl --workflow workflow.yml > workflow.full.drake
Output with default setup
I don't want to inlcude the whole file, but you get the idea
;
; Generated at: 2015-06-21T11:01:24
; This file was generated with the following options
; --workflow drake.yml
; --min 1
;
;
; Samples: test1, test2
;
;
; Starting Workflow
;
;
; Starting backup
;
;
; Variables
; Indir: /home/guests/jir2004/workflow
; Outdir: /home/guests/jir2004/workflow/output/backup
; Local Variables:
; INPUT: {$self->indir}/{$sample}.csv
; OUTPUT: {$self->outdir}/{$sample}.csv
; thing: other thing
;
/home/guests/jir2004/workflow/output/backup/$[SAMPLE].csv <- /home/guests/jir2004/workflow/$[SAMPLE].csv
cp $INPUT $OUTPUT
;
; Ending backup
;
;
; Starting grep_VARA
;
Run drake
drake --workflow workflow.full.drake
The following steps will be run, in order:
1: /home/user/workflow/output/backup/test1.csv <- /home/user/workflow/test1.csv [timestamped]
2: /home/user/workflow/output/backup/test2.csv <- /home/user/workflow/test2.csv [timestamped]
3: /home/user/workflow/output/grep_vara/test1.grep_VARA.csv <- /home/user/workflow/output/backup/test1.csv [projected timestamped]
4: /home/user/workflow/output/grep_vara/test2.grep_VARA.csv <- /home/user/workflow/output/backup/test2.csv [projected timestamped]
5: /home/user/workflow/output/grep_varb/test1.grep_VARA.grep_VARB.csv <- /home/user/workflow/output/grep_vara/test1.grep_VARA.csv [projected timestamped]
6: /home/user/workflow/output/grep_varb/test2.grep_VARA.grep_VARB.csv <- /home/user/workflow/output/grep_vara/test2.grep_VARA.csv [projected timestamped]
Confirm? [y/n] y
Running 6 steps with concurrence of 1...
--- 0. Running (timestamped): /home/user/workflow/output/backup/test1.csv <- /home/user/workflow/test1.csv
--- 0: /home/user/workflow/output/backup/test1.csv <- /home/user/workflow/test1.csv -> done in 0.02s
--- 1. Running (timestamped): /home/user/workflow/output/backup/test2.csv <- /home/user/workflow/test2.csv
--- 1: /home/user/workflow/output/backup/test2.csv <- /home/user/workflow/test2.csv -> done in 0.01s
--- 2. Running (timestamped): /home/user/workflow/output/grep_vara/test1.grep_VARA.csv <- /home/user/workflow/output/backup/test1.csv
Working on /home/user/workflow/output/backup/test1csv
--- 2: /home/user/workflow/output/grep_vara/test1.grep_VARA.csv <- /home/user/workflow/output/backup/test1.csv -> done in 0.01s
--- 3. Running (timestamped): /home/user/workflow/output/grep_vara/test2.grep_VARA.csv <- /home/user/workflow/output/backup/test2.csv
Working on /home/user/workflow/output/backup/test2csv
--- 3: /home/user/workflow/output/grep_vara/test2.grep_VARA.csv <- /home/user/workflow/output/backup/test2.csv -> done in 0.01s
--- 4. Running (timestamped): /home/user/workflow/output/grep_varb/test1.grep_VARA.grep_VARB.csv <- /home/user/workflow/output/grep_vara/test1.grep_VARA.csv
--- 4: /home/user/workflow/output/grep_varb/test1.grep_VARA.grep_VARB.csv <- /home/user/workflow/output/grep_vara/test1.grep_VARA.csv -> done in 0.01s
--- 5. Running (timestamped): /home/user/workflow/output/grep_varb/test2.grep_VARA.grep_VARB.csv <- /home/user/workflow/output/grep_vara/test2.grep_VARA.csv
--- 5: /home/user/workflow/output/grep_varb/test2.grep_VARA.grep_VARB.csv <- /home/user/workflow/output/grep_vara/test2.grep_VARA.csv -> done in 0.08s
Done (6 steps run).
Run in minified mode
As an alternative you can run this with the --min option, which instead of printing out each workflow prints out only one, and creates a run-workflow.sh which has all of your environmental variables.
This option is preferable if running on an HPC cluster with many nodes.
This WILL break with use of --resample, either local or global. You need to split up your workflows as opposed to using the --resample option.
biox-workflow.pl --workflow workflow.yml --min 1 > workflow.drake #This also creates the run-workflow.sh in the same directory
./run-workflow.sh
cat drake.log #Here is the log for the first run
2015-06-21 14:02:47,543 INFO Running 3 steps with concurrence of 1...
2015-06-21 14:02:47,568 INFO
2015-06-21 14:02:47,570 INFO --- 0. Running (timestamped): /home/user/workflow/output/backup/test1.csv <- /home/user/workflow/test1.csv
2015-06-21 14:02:47,592 INFO --- 0: /home/user/workflow/output/backup/test1.csv <- /home/user/workflow/test1.csv -> done in 0.02s
#So on and so forth
If you look in the example directory you will see a few png files, these are outputs of the drake workflow.
=cut
Acknowledgements
Before version 0.03
This module was originally developed at and for Weill Cornell Medical College in Qatar within ITS Advanced Computing Team. With approval from WCMC-Q, this information was generalized and put on github, for which the authors would like to express their gratitude.
As of version 0.03:
This modules continuing development is supported by NYU Abu Dhabi in the Center for Genomics and Systems Biology. With approval from NYUAD, this information was generalized and put on bitbucket, for which the authors would like to express their gratitude.
Inline Code Documentation
You shouldn't need these, but if you do here they are.
Attributes
full
Print the whole workflow hardcoded. This is the default
min
Print the workflow as 2 files.
Run the drake things
drake --vars "SAMPLE=$sample" --workflow/workflow.drake
workflow.drake
Our regular file
Subroutines
Subroutines
before run
Must initialize some variables
after get_samples
Things to do if we decide to do a min version
write_process
Fill in the template with the process
Ensure INPUT/OUTPUT exist
Prettyify the output a bit
AUTHOR
Jillian Rowe <jillian.e.rowe@gmail.com>
COPYRIGHT
Copyright 2015- Jillian Rowe
LICENSE
This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself.