How to predict the mature miRNA from its precursor

Question

[Beatriz]

Hello,

I am currently working on predicting ncRNAs using Infernal as my primary prediction tool. I have several datasets for different species containing microRNA precursors and their families, annotated based on Rfam. I am currently using RNAfold to predict the secondary structure and the MFE score to filter the results. However, to perform target prediction, I need to identify the mature miRNAs within the precursor sequences, and I am having difficulty finding suitable tools for this task. Can anyone provide me with some guidance on this matter?

Answer 1

[colindaven]

Can you get mature miRNA fastq data from your species of interest, trim it multiple times to get the expected length, then map it ? I think bowtie1 was a decent mapper for miRNA data back then, there might be better tools now.

Have a look for public data using a tool like https://sra-explorer.info/

[Beatriz]

I am working with non-model species and do not have information on their mature miRNAs specifically. However, I am considering downloading all the possible mature miRNA sequences from mammals (my group of interest) that can be derived from the precursors I've found and mapping them. Would this be a viable approach?

[colindaven]

If there's any data from closely related species I would consider that too. miRNA mapping is very difficult, as genomes are large and miRNAs are tiny - 21 bp or similar. I'm sure you know this though.

Remember to go for exact matches and not allow mismatches in your mappings to avoid false positives, at least initially.