How to quantify retrotranspon expression?

[DonDolowy]

Dear all,

I have a RNA-seq dataset (polyA-enriched, 2x50bp, 6-plexed) from mice. I have analyzed the data using both the Tuxedo pipeline, as well as the HTseq -> DESeq pipeline, where I have been mostly interested in protein-encoding genes.

However, now we are also interesting in quantifying expression of retrotransposons (IAP elements, SINEs, LINEs etc.) to see if there is any difference between wildtype and knockout.

This is the point where my self-tought bioinformatic skills come short. How would one go about this, since the mapped reads will not be associated with annotated genes?

I was thinking if I could extract the coordinates of the different elements from biomart and then use bedtools to somehow summarize their expression? Or can I just run cufflinks and use UCSC Repeatmasker track as annotation file?


Thanks a lot! Appreciate your inputs.

Best,

Kevin

[DonDolowy]

If anyone in the future has a similar analysis in mind, I ended up doing the analysis by mapping the reads to whole mm9 genome using Bowtie2. Convert the results to bed and then use bedtools intersect to intersect the results with the Repeatmasker annotation from UCSC. Sum up the counts followed by generation of counts matrix that can be used for DESeq.