Splitting the Spliced reads from genome Alignment

Question

[EVR]

Hi,

I have a RNA-seq reads mapped onto genome using tophat2 and obtained a "bam" file which is later converted into a BED file. I extracted only the relevant columns from bed file which includes the Scaffold, start, end, orientation and CIGAR. For spliced read liek following

scaffold8075    66972   68644   -   48M22N2M

The CIGAR string says that the first 48 bp of read matches to genome, then 22 bp of introns and later the last 2 bp of the read matches. Now I would split the above read based on matched regions that matches like following

scaffold8075    66972   67020   -   48M22N2M
scaffold8075    68642   68644   -   48M22N2M

In simple words, I would like split the spliced reads based on its exon-exon alignment. Any guidance would be appreciated. Thanks in advance.

[WouterDeCoster]

Have you read the options of bedtools bamtobed? Looks like -split is what you need.