I have some long non coding RNA sequences for which I would like to predict whether there any sites for the binding to specific miRNAs. I already tried using RNA hybrid (http://bibiserv.techfak.uni-bielefeld.de/rnahybrid/welcome.html) and DIANA (http://diana.cslab.ece.ntua.gr/), but unfortunately without success. With RNA hybrid I managed to have some nice pictures showing the actual 2 dimensional structure formed by the hybridization between the lncRNA and the miRNA (see attached picture ; red is lncRNA, green is miRNA and Mfe is the minimum free energy of hybridization) but I am not too sure if I can trust these kind of predictions, because the database managed to predict some binding sites for all my lncRNAs and not only on the few ones where I expected the binding sites to really be.
I could not do any analysis on DIANA because it does not accept my ensemble lncRNA annotations nor the genomic locations for the lncRNA that are not annotated.
Can anybody help me with this? Can you maybe suggest another software/database which I could use?
Hi Paolo
Those utilities always depend on other databases. If you have the sequence of the lincRNA and the miRNAs that you want to test, I would advice you to use Miranda. This is a pretty classic tool, but the main advantage is that you can install it locally in your computer and do predictions very quickly. You will need a Fasta formatted file with the lincRNAs and other with the miRNA sequences.
Here is the link to download the software
http://www.microrna.org/microrna/getDownloads.do
The software is in the source code, and you must compile it, but this is pretty straightforward in Linux.
Hope it helps
Best. Paco
Dear Paco,
Thank you very much for your reply. Thank you for posting here the link to the Miranda software. I am just a bit confused about which link I am supposed to use? Also are you sure that this is a normal software/programme that you have to install on your computer?
Regards,
Paolo
Paolo...
This is the link for the most recent software
http://cbio.mskcc.org/microrna_data/miRanda-aug2010.tar.gz
The other links are for databases.
Best. Paco
You should definitively try jMiRanda.jar
Using this downloadable application, you can insert both the mature miRNA sequence and the target sequence (I had a complete cDNA of a gene, so I think you can enter whatever target you want), and it predict up to 8 possible nucleation sites. You then have to chose preferentially those with the higher (more negative) DeltaG, according also to the matching sites.
Hello Paolo,
I guess you can use following approach:
a) You can Download miranda from http://www.microrna.org/microrna/getDownloads.do
b) Download all micrornas fasta from human and input as query sequences
c) Give Long non coding RNA's as target and run miranda
You get all possible binding locations of available micrornas on your long RNA's
Some of the latest lncRNA databases offer insight into miR binding sites. Try StarBase at http://starbase.sysu.edu.cn or DIANA-LncBase at http://diana.imis.athena-innovation.gr/DianaTools/index.php?r=lncPredicted/index
There are other places to look too; just google!
StarBase is off for more than two weeks! Diana didn't allow you to download the data. Where can one find another place for lnCRNA:miRNA interaction database?
Hello. Yes, I have also noted that StarBase is not working. Does anyone know why? Or when/whether it will work again? Thanks!
- Badges
- Science method
- Similar topics
- Biological Science
- Genetics
- Genetic Analysis
- genetic techniques
- Sequencing