I have Illumina data of 1 x50bp from small RNA library. I am interested in identifying known and novel miRNAs in my plant, whose reference genome is not available. Here first I want to count the reads aligning to non-coding rRNAs like tRNA, rRNA, snoRNA, snRNA, siRNA etc except miRNAs. After removing those reads, I would like to align remaining reads with miRBase using blastn or bowtie to identify known miRNAs. Now problem is that, I want to use Rfam for detecting reads of different RNA class, thus I concatenated the fasta files of 2686 families (2,487,655 sequences) but Rfam v13 sequences does not have proper information of the class they represent in their header. So my concern is how can I count and extract the reads mapping to different classes of RNAs using Rfam.