I have done de novo assembly of pair end raw read sequences and resulted transcripts sequences were separated based on coding potential into two categories: long non-coding RNA transcripts and coding RNA transcripts.
I want to find how many of long non coding can make hairpin loop structure and how can I annotate these miRNA which are generated from long non coding RNA in plants.
For the first question: "how many of long non coding can make hairpin loop structure?"
The easiest thing to would be to run your sequences through an RNA secondary structure prediction tool. There's a few tools for doing this but the most commonly used are RNAfold from the ViennaRNA package and mfold.
This will give you an output that looks like this, where the parentheses indicate which bases are paired with each other.
GGGCUAUUAGCUCAGUUGGUUAGAGCGCACCCCUGAUAAGGGUGAGGUCGCUGAUUCGAAUUCAGCAUAGCCCA
(((((((..((((.........)))).(((((.......))))).....(((((.......)))))))))))).
To see if there's hairpins, you'll need to annotate the elements within it. Take a look at this question on biostars for an example of how to do this. In short, you can use the forgi library to produce an annotation like this:
(((((((..((((.........)))).(((((.......))))).....(((((.......)))))))))))).
sssssssmmsssshhhhhhhhhssssmssssshhhhhhhsssssmmmmmssssshhhhhhhsssssssssssse
If there's an h in the output annotation, you have a hairpin. The remainder of your question about annotating miRNAs will be left to somebody else.
(.), (..), (...)).
$\endgroup$