We have done piRNA-Seq, and for data analysis we meet some trouble. The wet lab pipeline:

1. We focus on three proteins and we want to known if those proteins can specifically interact with piRNAs.
2. tissue came from mouse testis were used for RNA immunoprecipitation.
3. after RIP, piRAN was selected by gel selection, and cut 20nt~40nt RNA for small RNA library construction
4. single end 50nt strategy was used for piRNA sequencing

Data analysis pipeline:

1. cutadapt was used for raw data clean,remove adaptor, filter low quality data
2. clean data have been mapped to mouse genome mm9 by bowtie with 0 mismatch and only unique mapped reads were accepted for next step analysis. But I still doubt piRNA may be come from repeat element in genome, and they could mapped to multiple position.
3. then I used htseq-count and DESeq to compare differential enriched piRAN.GTF file for piRNA download form piRANBank and feed for htseq-count. We have sequenced IgG and Input control, and both of them enriched many piRNA.

Any one who knows if we can use this pipeline for piRAN analysis? And is there a better method?