[question] how to get miRNA sequences for target prediction

[juancresc]

I want to run a target search using a program like https://github.com/carringtonlab/TargetFinder to find putative target genes. How can I get miRNA sequences? I see in the results file the MajorRNA sequence, but I think it would be best to use 20/24 nt miRNAs. Or what would you recommend for target prediction? Thank you

[MikeAxtell]

Hi Juan, thanks for your question.

The 'MajorRNA' in the 'Results.txt' output is the numerically most abundant small RNA aligned to the locus. For MIRNA loci, in general that should be the functional mature miRNA. Nearly all true plant mature miRNAs are 21nts, some are 20, and some are 22. Other sizes are very very rare.

Hope this helps

On Fri, Jun 29, 2018 at 9:03 AM Juan Manuel Crescente < notifications@github.com> wrote:

I want to run a target search using a program like https://github.com/carringtonlab/TargetFinder to find putative target genes. How can I get miRNA sequences? I see in the results file the MajorRNA sequence, but I think it would be best to use 20/24 nt miRNAs. Or what would you recommend for target prediction? Thank you

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/MikeAxtell/ShortStack/issues/75, or mute the thread https://github.com/notifications/unsubscribe-auth/AGiXiffkSJiwNmd4MkY3DepbGm-pXmFMks5uBiWHgaJpZM4U8_J4 .

-- Michael J. Axtell, Ph.D. Professor of Biology Penn State University http://sites.psu.edu/axtell

[juancresc]

Hi Dr Axtell, thank you for your answer.

You think we should use only miRNA clusters flagged as 'Y' for target prediction? In my experiment there are 137 miRNA but only 6 of them are differentialy expressed, and we are only interested in those.

I'm thinking N4, since Major RNA length is in the range. Maybe there are a lot of false positives, but for a genome-wide overview that could be OK.

BTW Shortstack is very comprehensive, and discussions here are really helpful, thanks!

[MikeAxtell]

I would not assume everything N4 and above is a microRNA. You could do some other filtering though. Such as, only take 'Major RNAs' that are 20,21,or 22 nts in length, and which start with a 5'-U. The big concern is using 24 mers are queries for target prediction, since nearly all 24mers are RDDM related, and there is scant evidence that RDDM-related small RNAs target mRNAs in trans.

Thanks for your comments and I'm glad you are finding the script useful for your research !

Mike

On Mon, Jul 2, 2018 at 9:31 AM Juan Manuel Crescente < notifications@github.com> wrote:

Hi Dr Axtell, thank you for your answer.

You think we should use only miRNA clusters flagged as 'Y' for target prediction? In my experiment there are 137 miRNA but only 6 of them are differentialy expressed, and we are only interested in those.

I'm thinking N4, since Major RNA length is in the range. Maybe there are a lot of false positives, but for a genome-wide overview that could be OK.

BTW Shortstack is very comprehensive, and discussions here are really helpful, thanks!

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/MikeAxtell/ShortStack/issues/75#issuecomment-401805022, or mute the thread https://github.com/notifications/unsubscribe-auth/AGiXifU3AVXierFp2udGEVrSElgL9lhXks5uCiC8gaJpZM4U8_J4 .

-- Michael J. Axtell, Ph.D. Professor of Biology Penn State University http://sites.psu.edu/axtell

[juancresc]

I've noticed that start with a 5'-U is not in the MIRNA analysis codes filters