How to compare single-end vs paired-end RNA-seq data?

[mick]

Hi all,

I’d like to compare RNA-seq data of WT (single-end RNA-seq) vs KO strain(paired-end RNA-seq), as follows:

wt 0hr 1hr 3hr (single-end)
KO 0hr 1hr 3hr (paired-end)


Since I’m interested in the differentially expressed genes during time course between wt and KO strains, I thought by using bedtools multicov option, I can compare these data.

I used the following commands;

bedtools multicov -s -bams wt_0hr.bam wt_1hr.bam wt_3hr.bam -bed gene.bed > wt0-1-3hr.txt
bedtools multicov -s -bams KO_0hr.bam KO_1hr.bam KO_3hr.bam -p -bed gene.bed > KO0-1-3hr.txt

(If options were not correct, please let me know.)

My questions are...

＊How does bedtools multicov treat paired-end reads? I mean, if both ends of a paired-end read were mapped to the same gene in the BED file are they then counted once or twice?

＊If comparing same condition data (= same sample preparation method), is it OK to compare the read count calculated from bed tools multicov?(Is there any need of normalizing data, like FPKM/RPKM or something?)


Thanks in advance.