Pirnas Sequence Files In Fasta Format For Rat And Mapping Small Rna Seq Reads

Question

[tjcstockholm]

I would like to search in my small-rna sequence set for pirnas and mirna. I managed to quantify the mirnas from the small-rna sequences.

I would like to know if there is any tool to find or predict the pirnas in the small-rna sequence dataset? Also, Is there any repository for pirnas where I can download the pirnas for rat in fasta format?

Any information will be helpful.

Thanks in advance

TJC

Answer 1

[Martombo]

You can find a database of known piRNA sequences here

if you're trying to map reads/sequences from a NGS experiment on piRNA databases you could find some difficulties: piRNA sequences are very similar to transposon sequences, so if you have transposon sequences in your data they can be misinterpreted as piRNA. to solve this problem, remember to use only reads of length around 23-29. also, try to focus on reads uniquely mapped on the genome. this can remove quite a bit of piRNA sequences (10-20%) but it definitely helps removing a lot of transposon sequences. another useful thing to do is to focus on known piRNA clusters for your organism (information also present on pirnabank). I have some experience of working with piRNAs in drosophila, so all the tips here come from a study in such organism. in other species it could be different.

A reference

Answer 2

[geneart$$]

I have done some similar work with my small RNA sequences. I used proTRAC software to identify any piRNA clusters. proTRAC is developed by Dr.David Rosenkranz's group. THis can be run on command line or as a user interface. I have used command line options The pdf for the user documentation mentions , to use a probabilistic option if the samples are not selectively enriched for piRNA sequences (martombo's suggestion here might help !) and has contamination with other tRNAs etc.

My samples were generated to exclusively enriching for miRNAs but also see fragments of tRNAs etc as the source is from exosome ! However I just ran the default parameters and still obtained clusters. Although my analysis is not complete I was excited to introduce this software on this forum as proTRAC uses the information on IBAB, and other traditional piRNA sources and has shown to work better in predicting clusters that IBAB missed. Something worth checking out: http://sourceforge.net/projects/protrac/

I would appreciate any inputs on the use of proTRAC if anyone has had the opportunity to use it!