Usually, algorithms predict miRNA targets with less than 50% accuracy. This is not science. I would like to isolate all the miRNAs based on their physical association with my mRNA of interest. I though of using synthetic 3'UTR promoter fused to GFP or else for immunopreciptation thereafter but what about non 3'UTR miRNAs? Any suggestions that don't cost an arm?
you can also use miRWalk to predict miRNAs for your target gene/pathway and vice versa,
http://www.umm.uni-heidelberg.de/apps/zmf/mirwalk/
but this is mere prediction and you need to know whether the respective miRNA is expressing in your target cells or not for which you need to go for deep sequencing.
Hi, I do agree with Darrell Green: a "blind" and "ab initio" approach could be quite impossible in practice. I also agree with you when you say that algorithms predict miRNA targets with less than 50% accuracy are a bit disappointing, but I suggest you to actually use their results as a method to actually filter out miRs that are really unlikely to bind to your target. These algorithms are based on biological rules: a miR must recognize its target by using its sequence, so if your target lacks of sequences that could be used as binding sites a number of miRs you just don't have to investigate more on them.
In my opinion techniques that uses a "fishing" approach to purify miRs using an mRNA as a bait in WT cells are very prone to false negatives because they strongly depends on the concentration of microRNAs in your culturing conditions.
On the other hand HT luciferase assays approaches may produce a number of physiologically irrelevant miR-mRNA pairs, but I think they provide much more information.
Cheers.
- Badges
- Science topic
- Similar topics
- Biological Science
- Biochemistry
- Biomolecules
- Nucleic Acids
- RNA
- Small RNA
- microRNA