Tweet
Hello,
Does anyone have any information/advice on whether strand information from aligned reads is taken into account during the identification of de novo differentially methylated regions (DMRs)?
I have used several different DMR identification software (BiSeq, DSS and eDMR) and all produce a list of candidate regions with their respective genomic coordinates but no strand information ("*"). Is the idea here that methylation within a identified DMR likely impacts regulation of genes in close proximity irrespective of strand?
In downstream analysis this lack of strand information appears to cause issues with the annotation of DMRs. I have used the GenomicRanges package to identify gene regions that overlap with specific DMRs and then used this information in GO term enrichment analysis. However, some of the genes included as overlap hits do not seem to have any aligned CpGs, suggesting that the differentially methylated CpGs that form the overlapping DMR were located on the alternate strand.
If this is the case, does anyone have any advice on how best to annotate DMRs where the strand information is missing to prevent these errors or maybe it is an irrelevancy?
Thanks!
Does anyone have any information/advice on whether strand information from aligned reads is taken into account during the identification of de novo differentially methylated regions (DMRs)?
I have used several different DMR identification software (BiSeq, DSS and eDMR) and all produce a list of candidate regions with their respective genomic coordinates but no strand information ("*"). Is the idea here that methylation within a identified DMR likely impacts regulation of genes in close proximity irrespective of strand?
In downstream analysis this lack of strand information appears to cause issues with the annotation of DMRs. I have used the GenomicRanges package to identify gene regions that overlap with specific DMRs and then used this information in GO term enrichment analysis. However, some of the genes included as overlap hits do not seem to have any aligned CpGs, suggesting that the differentially methylated CpGs that form the overlapping DMR were located on the alternate strand.
If this is the case, does anyone have any advice on how best to annotate DMRs where the strand information is missing to prevent these errors or maybe it is an irrelevancy?
Thanks!