Hello,
I have 100 miRNAs. I want to find their target mRNAs using TargetScan, miRDB. I can search miRNA-mRNA interaction in this tools one by one. Is there a practice way to find 100 miRNA-mRNA interactions in one search using these tools. Thanks
We generally do this by bulk downloading the target match files.
There are several ways to do this.
You can download the conserved targets of conserved miRNAs from the targetscan website here: http://www.targetscan.org/vert_72/vert_72_data_download/Predicted_Targets_Info.default_predictions.txt.zip
Alternatively the ENCORI database (although the site is actaully not responding for me right at this moment) has target predictions for many prediction algorithms, along with experimental evidence from AGO-CLIP experiments available to download as a text file.
You can then use awk, R, pandas, or even excel to filter these files to the miRNAs you are interested in.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
Thank you very much i.sudbery. I will try with R. I cannot access ENCORI database from our country. Thanks again
Why do you use only conserved targets of conserved miRNAs? Can we use all of them (poorly conserved / conserved) for searching? Thanks
Conservation is a signal of function - if a site is functionally improtant, it must surely be under purifying selection, because if it is important, then breaking it must lead to less fitness.
It is of course possible for a site to be important in one species, but not others. Those sites are likely constrained within the specifies, but not between species. These sites will be missed by only looking for conserved sites.
However, we know that by far the largest majority of binding sites predicted by sequence alone are not functional. Until we have a good feeling for what makes a binding site functional or not (which I don't think we do at the moment), conservation acts as a way of deciding which are more likely to be real, and which not.
I downloaded the files from TargetScan, miRDB and mirtarbase and searched 100 miRNAs / their target genes in these files. Although all of these 100 miRNAs / target genes could found in miRDB and mirtarbase files, not found in TargetScan file. Only 3 miRNAs and their target genes could found.
Is it because TargetScan contains only conserved miRNAs/mRNAs and others both conserved and also poorly conserved miRNAs/mRNAs? In this case, if I retrieve overlapping mRNAs in all three databases, I can use only 3 miRNAs and their target genes.
Yes, its almost certainly because the TargetScan file above has only the conserved targets of conserved miRNAs. The TargetScan Website has files for non-conserved miRNAs and non-conserved targets as well.