Mirna Downregulates Mutated Gene Responsible For A Certain Disease?

Question

[Agatha]

I am using a tool for predicting miRNA binding sites and I am currently analyzing mRNA microarray expression data from failing hearts (hypertrophic cardiomyopathy) versus normal .

Question:

 Considering that I have found a potential miRNA that might be involved in gene regulation,
     and I have used a target prediction tool to look for potential targets
        and One of the highly scoring target genes(Kras) is implicated in the disease (mutation)

what can be inferred about this miRNA-mRNA binding?

Answer 1

[JC]

Well, I first will look for a few things:

1. Do you have anti-correlation for the miRNA-mRNA in terms of expression? Not only on heart, in other tissues/conditions.
2. How well defined is the bindig site? entropy of the match, conservation in other species, permutation tests, ...
3. There are previous reports for this miRNA that provide more information in the disease.

From there, you need to validate in the lab the prediction, there are several ways to see if an interaction really exists and how strong.

[Agatha]

1. No correlation.I am looking for potential miRNA signatures in mRNA expression data.
2. The binding site: sequence complementarity, sequence- word overrepresentation across the mRNA sequences, transcript expression change
3. There are no previous reports regarding a bunch of miRNAs which have very significant binding sites. However, I could find reports about the predicted target genes being up or downregulated in the specific disease

[Agatha]

so I was wondering how correct would it be to consider that the specific miRNA might have influence the development of the hypertrophy by binding to a transcript of a gene who is responsible for that disease - if mutated

[JC]

it's a good hypothesis, but I think you need more work to filter the false positives and some lab validation

[Sudeep]

I support what JC just said, and AFAIK almost all the miRNA binding prediction algorithms are just based on 3' sequence complementarity, so you might be needing some lab validation to prove that miRNA binds to your gene of interest and this particular miRNA is expressed in your samples.