Questions About Using Aggregated Results from MPRAsnakeflow for Allelic Comparison

[biozzq]

Dear all,

I recently ran the MPRAsnakeflow pipeline and obtained the aggregated results from the file reporter_experiment.oligo.XXX.fromFile.default.min_oligo_threshold_10.tsv.gz. My goal is to compare the effects of different alleles on the activity of candidate regulatory sequences. Specifically, I would like to use the dna_count and rna_count generated from this file for downstream analysis. Could you please confirm if it is appropriate to directly use these aggregated results for comparing allelic effects?

library(mpra)
dna_data <- read.table("dna_count",head=T,row.names=1)
rna_data <- read.table("rna_count",head=T,row.names=1)
mpraset <- MPRASet(DNA = dna_data, RNA = rna_data, eid = rownames(dna_data), eseq = NULL, barcode = NULL)
design <- data.frame(intcpt = 1, mutant = grepl("mutant", colnames(mpraset)))
block_vector <- rep(1:3, 2)
mpralm_allele_fit <- mpralm(object = mpraset, design = design, aggregate = "none", normalize = TRUE, block = block_vector, model_type = "corr_groups", plot = TRUE)
toptab_allele <- topTable(mpralm_allele_fit, coef = 2, number = Inf)
toptab_allele

The partial contents of dna_count and rna_count are as follows:

I am currently using version mpra_1.6.0. Could you please advise on how to export the normalized DNA and RNA count results?

Thank you very much for your time and assistance. I look forward to your reply.

Best wishes,
Zheng zhuqing

[lmyint]

It is appropriate to use pre-aggregated counts for your analysis, as you have done.

To get the normalized DNA and RNA counts, you can use the normalize_counts() function directly as in normalize_counts(mpraset, block = block_vector). This creates a new MPRASet object with normalized counts, on which you can call getDNA() and getRNA() to obtain those normalized counts.