4.5 years ago by
Your link is incomplete, I dont know which study you are referring to exactly but still!
Your steps could be:
1. Pulling the sequences downs (whole ncRNA), mapping them to respective genome, read filtering etc and obtain a final mapped dataset (Tophat/Bowtie can do that).
- Generate a reference ncRNA 5'/3' regions and check which fragments map to them. This will give you your answer.
You could also pull down ncRNA reference fasta (I dont know where exactly from), generate 5'/3' list, index them using Bowtie and map your whole dataset to that refence. So, essentially you would have a BAM file which you could convert to fasta/BED, depending on what you want and gives you your answer.
UCSC Table Browser should have the references you wanted or try biomart martview.