miRNA Target Site prediction from unpaired data

Question

[hAjmal]

Hello Everyone, I am a computer scientist and started working with a group of researchers working in bioinformatics.I have little or no background of genomics. What I have been asked to do is,I have a set of miRNA expression analysis data, about 40 patients divided into 6 cohorts, including a control group. miRNA expression levels of each group of pateints are measured against teh control group and logFC and p-values are present. What I have been asked is to group the miRNA (up or down regulated) into miRNA families or GO groups. and then look to see are those target genes up or down regulated the publicly available data sets. That is, first determine the dysregulated miRNA in our data set and then correlate them with their expression levels (the target mRNA) from these data sets.

Now so far I understand is that negative logFC is for down regulation and positive is for up regulation. If I pick out say, first 20 miRNA from the miRNA data set on basis of their lower p-values, how do I map those miRNA to their targets in publicly available data sets. I have read a few papers including 'Integrative Approaches for microRNA Target Prediction: Combining Sequence Information and the Paired mRNA and miRNA Expression Profiles(http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3583062/#R39)'. And other predcition tools like GenMir++. But they require that the mRNA and miRNA data be paired that is obtained from the same samples.This is not our case clearly. We dont have mRNA data from the same samples. Do I use online tools like targetScanS or Tarbase to map miRNA to mRNA?

Advise would be appreciated. Thanks

Answer 1

[BioRyder]

You can use microRNA.org to see the list of miRNA target if you are having list of miRNA (Human).This application is online and does not need mRNA or UTR region to predict the miRNA target reads. miranda is an offline tool which used for miRNA target prediction.In miranda you have to provide two fasta file as input, one is your miRNA sequence file and another is mRNA or UTR region of your organism(in your case human)

[hAjmal]

Thank you. I will look into these. What about TarBase? I have tried it in TarBase, some mRNA were identified but there were no experimentally supported targets for a few of the miRNAs.