Automate Illumina GenomeStudio With Python Script

Question

[travis.m.couture]

Hey everyone. I am currently working on automating a SNP pipeline and need some help figuring out how to interface Python with GenomeStudio. I want to be able to pass input from a python script into GenomeStudio so that the entire analysis process can be automated and manual intervention is only required rarely. Does anyone have any experience with this? I've done a few Google searches and haven't come up with anything.

Thanks, Travis Couture

[Steven Lakin]

I've been in this scenario before, and if you have the choice of language, I'd recommend utilizing bash shell scripts (Unix/Linux) to create a pipeline workflow for your application. Python is notoriously difficult to use in multi-processing or alternative process applications, both since it has a global interpreter lock for multi-threading and also because Python is not usually the most efficient language for subprocessing.

That being said, here is the module you will have to use if you utilize Python: https://docs.python.org/3/library/subprocess.html

You'll likely also be familiar with the multiprocessing module.

[travis.m.couture]

Unfortunately I'm working in a Windows environment for this project. Could the same approach for bash be taken MS-DOS? If so where would I start. If not I will have to read up on the multiprocessing module. I am familiar with he subprocess module.

[travis.m.couture]

I called Illumina and they were no help. I am still researching this but I'm hoping someone out there can give me some direction.

[Jonathan King]

Apologies for my vagueness - I don't have experience with your exact situation, but here are my thoughts:

I've used Python for this, but only in combination with other bash scripts on a Unix based OS. Python's subprocess methods work fine, but it's surely easier if you find a way to create and run bash scripts. There are several resources (though I don't have experience with them personally) for emulating bash on Windows if you decide to go that route.

Also, what kind of input does GenomeStudio expect? Theoretically, if the second application in the pipeline can read from stdin, simply have Python print the required input (this goes to stdout) and the downstream application can capture this.

Best of luck!

Answer 1

[blrhgeruser1]

Not exactly automating GenomeStudio, but what you could do is turn on AutoConvert on the iScan instrument and have it generate gtc files for each sample in your experiment. Then you can use the recently published open-source Python gtc file parser from Illumina to write a script for generating reports or whatever else you want to do with the data: https://github.com/Illumina/BeadArrayFiles. In the examples folder of that repo, there's an example script that lets you generate final report files from gtc files, thereby letting you avoid having to use GenomeStudio at all.

Or if you're working with idat files instead, what you could do is install and run AutoConvert off the instrument and write some Python code to go from idat + bpm + egt --AutoConvert--> gtc --python gtc file parser--> reports

[travis.m.couture]

I didn't know about this. Thanks for the info! We'll give it a try.

Answer 2

[Giulio Genovese]

Nowadays all of this can be done on the command line without using GenomeStudio as explained here

[Ginevra]

Hi! I am wondering if the gtc2vcftool can help me to get a .txt report with the LRR and BAF info (similar to the one produced using Genomestudio).