miCLIP2-DMA-source-code

Author: You Zhou, Kathi Zarnack

A function for running the bin-based differential methylation analysis (DMA)

Installation

install.packages("path_to/miCLIP2SourceCode_1.0.1.tar.gz", repos = NULL, type="source")

Description

In order to learn about the features of genuine m6A sites in the miCLIP2 data, we sought to extract all miCLIP2 peaks that significantly changed in the Mettl3 KO mESCs. However, changes at individual peaks were overshadowed by massive shifts in gene expression in Mettl3 KO cells, with more than 2,809 genes altered at least 2-fold in comparison to WT mESCs (false discovery rate [FDR] ≤ 0.01). These massive shifts in the underlying transcript abundances meant that miCLIP2 read counts at individual peaks could not be compared directly. In order to overcome this shortcoming, we tested several strategies for differential methylation analysis to account for the substantial gene expression changes in the Mettl3 KO cells. Best performance was achieved with the bin-based approach, here we provide a function to do the bin-based differential methylation analysis for any miCLIP2 data.

Workflow for running the binBased_DMA function

The binBased_DMA function requires the gene counting result by htseq-count
and the single nucleotide peaks that output by the pureCLIP with the truncation signal miCLIP2 data as input.

After user collecting the output from the htseq-count and pureCLIP, user can follow the following workflow to complete the bin-based differential methylation analysis.

1. Import the bed file of the peaks as a GRanges object.
2. Assign the truncation signal to the peaks. For doing that, one option could be the function truncationAssignment in the package m6Aboost.
3. Generate the TxDb annotation object for the experiment.
4. Fill the parameters and run the binBased_DMA function.
5. The binBased_DMA exports a GRanges object that contains the result of the peak changes.

Example of usage:

path2htseq <- "/path/to/the/htseqresult/"
conditionS <- c(rep("KO",3), rep("WT",3))
condition1 <- c("read_WT_1", "read_WT_2", "read_WT_3")
condition2 <- c("read_KO_1", "read_KO_2", "read_KO_3")

result <- binBased_DMA(peaks, path2htseq, conditionS, txDB=NA, condition1=condition1, condition2=condition2)