search

daisybio / ASimulatoR

ASimulatoR: splice-aware RNA-seq data simulation https://doi.org/10.1093/bioinformatics/btab142
GNU Lesser General Public License v3.0
6 stars 1 forks source link

two groups with distinct prob per group per event? #5

Open ls233 opened 7 months ago

ls233 commented 7 months ago

Hey guys, thanks for the tool! Very useful!

I'm a bit confused about the "num_reps" parameter. And also who it relates to the event probabilities.

I have implemented an alternative splicing (AS) analysis pipeline and I'd like to test validness of the my AS detection and abundance (PSI) using a synthetic dataset.

Generally I'd to create a simulated dataset for two groups with 2 samples each. Specifically, I'd like to simulate RNAseq reads for two groups in such a way that one group would have a higher abundance for one of the isoforms. Having two/three samples per group would allow to test for significant difference across the two groups.

quirinmanz commented 7 months ago

Hi,

num_reps = c(2,2) creates 2 groups with 2 samples each. Note that isoform switches are not controlled at the moment. See also #4 for more information.

If you have more questions, just comment here.

Best, Quirin

ls233 commented 7 months ago

I looked over comment #4 but I'm not sure that it addressed my question. Specifically, I'd like to simulate RNAseq reads for two groups (few samples each) in such a way that one group would have a higher abundance for one of the isoforms. I don't need to control the events themselves but the abundance of the participating isoforms.

quirinmanz commented 7 months ago

ASimulatoR was created to benchmark event detection tools. If I understand correctly, you are analyzing differential splicing and isoform switching.

[...]

You could still create gtfs with splice events using ASimulatoR and then give this custom gtf to polyester with your own fold_change table. This is not recommended, but should work.

Do I understand correctly? Is this what you want to do?