Running rMATS on sequencing files from different machines

[Shiri Jaffe]

Hi,

I would like to run rMATS on sequencing files from two different sequencing platforms: 

1. NovaseqX as 161*161 bp

2. Novaseq6000 as 161*161 bp

I have 3 groups of files:
WT- 5 from Novaseq6000, 1 from NovaseqX

Mutant- 5 from Novaseq6000  , 1 from NovaseqX
Heterozygous- 5 from NovaseqX  , 1 from Novaseq6000

Can I use rMATS to compare splicing changes between the three groups? Are there any adjustments I should apply before running rMATS due to the different platforms?

Thanks

[kutsc...@gmail.com]

rMATS can be used for pairwise comparison of three groups: https://github.com/Xinglab/rmats-turbo/tree/v4.3.0#running-the-statistical-model-separately

I don't think you'll need to make any adjustments for data from those two platforms. It seems like all the data is paired 161 bp. If some of the data had a different read length then you could run separate prep steps with the parameters set as needed for each sample: https://github.com/Xinglab/rmats-turbo/issues/527#issuecomment-3456287054

Eric

[kutsc...@gmail.com]

The significant results might be caused by a batch effect. One thing you could try is making a PCA plot of all the samples by InclusionLevel. If the samples cluster by when they were sequenced rather than by their genotype then that suggests a batch effect

These links have discussions of batch effect:
https://github.com/Xinglab/rmats-turbo/issues/111
https://groups.google.com/g/rmats-user-group/c/bjhADtjg3iU/m/EvjvS5SZAwAJ

This link has code for making a PCA plot: https://github.com/Xinglab/rmats-turbo/issues/398#issuecomment-2391636164

You could also check if certain samples have a lot of reads filtered out for some reason. rMATS will print a section starting with "read outcome totals across all BAMs" that shows counts of reads that were filtered for various reasons. There is also a _read_outcomes_by_bam.txt file in the --tmp directory

Eric