mae pipeline not usable - too slow about drop HOT 5 CLOSED

Hi, we usually run the MAE pipeline in vcf files generated from WES. The ones from GTEx are generated from WGS, which makes the process much slower.
DROP uses the function to count alleles from GATK, which is the one taking several hours.
Do you need to run MAE on the GTEx samples? For this module, you do not need other samples as controls because the statistical test is done per sample, not as a group.

from drop.

Oh ok. I will run it only on my samples then.
How do I do this? There are two settings for mae: 'groups' and 'qcGroups'.
What is the difference? Which samples should be in each type of group?

from drop.

You can specify in the DROP_GROUP column in the sample annotation which are your samples, and then write that group in the mae groups parameter. A sample can belong to multiple groups, which are comma-separated in the DROP_GROUP column.
The mae 'groups' specify the groups that are going to be tested for mono-allelic expression.
The qc module checks whether a DNA and RNA combination actually match. So the qcGroups specify the groups that you want to test this. In your case you should assign the same value to both.

from drop.

Thanks. What is the difference between the qc module of mae versus the snakemake sampleQC command?

from drop.

They're the same.
If you execute snakemake mae, both the mae and qc are executed. If you execute snakemake sampleQC, only the qc is executed.

from drop.