gravatar for 2822462298

2 hours ago by

Hi all,
I am using miRDeep2 to analyze miRNAs from a non-model organism (a butterfly) without any mature miRNAs or precursors in miRBase. First, I used to map the pre-filtered sRNA-seq reads to the genome: ${i}.fq -e -h -l 16 -m -p genome_index -s ${i}_read.fa -t ${i}_map.arf -v -o 12

Then, I used to predict precursors: ${i}_read.fa genome.fa ${i}_map.arf none Drosophila+Lepidoptera_mature.fasta none 2>${i}_report.log

The mature and hairpin reference data were set as "none" and I used all lepidoptera+drosophila mature sequences downloaded from miRBase as the input of 'related species'. I got a list of predicted miRNAs in the csv and also the html files. Here's the problem:

Three fields are blank in my html files:

"estimated probability that the miRNA candidate is a true positive"

"rfam alert"

"miRBase miRNA"

Based on this I have no idea if my miRNAs are conserved (identical mature sequences or hairpins can be found in other/related species) or novel. Is it because I did not include any reference data of mature or hairpin? Since miRNAs have not yet been investigated in this species, can I use ALL mature and hairpin available in miRBase as my reference to see if a match can be found for each individual miRNA? In that case: ${i}_read.fa genome.fa ${i}_map.arf miRBase_all_mature.fasta Drosophila+Lepidoptera_mature.fasta miRBase_all_hairpin.fasta 2>${i}_report.log

Would this work for my purposes?

Source link