Predicting Cancer-related MicroRNA Target Genes with Data Fusion Method

MicroRNAs (miRNAs) are a class of small non-coding RNAs that bind to their target mRNA sequence and induce either translation repression or mRNA degradation.Recent studies have indicated that miRNA may play an oncogenic or tumor suppressor role in tumorigenesis. When miRNA plays an oncogenic role, it targets tumor suppressor gene (TSG) and leads to tumor formation. On the other hand, there are reports which show that miRNAs are repressed in human cancers. In other words, if miRNA plays the role of tumor suppressor due to deregulation of miRNA, oncogene (OCG) may be over-expressed and induce tumor formation. To study this hypothesis, it is important to identify cancer-related miRNA-mRNA pairs. In this study, we put together several miRNA target gene prediction tools, i.e. TargetScans, RNAHybrid and PicTar, to facilitate such search. An integrated approach is employed, because each prediction tool is supplement to each other.


Threshold for each target prediction tool,

TargetScans context score < 0, RNAHybird p-value < 0.05, PicTar score > 0, and miRanda z-score > 3.

Here a voting score is assigned for each query based on the number of significant predictions. Given four miRNA target prediction tools, voting score ranges

from 0 to 4. A higher voting score implies a higher degree of prediction consistency.

In case of selecting the "All tools" option, it is recommend that you set a score value >=3.

Service 1 - select a OCG, TSG or cancer-related gene, this service allows you to search for the predicted microRNA.

Tool Type GeneName Refseq ALL Tools : score =

Service 2 - select a gene (type: OCG, TSG, other) and microRNA gene, this service allows you to search for the predicted target genes.

Tool Type miRNA ALL Tools : score =

Service 3 - select a threshold for the "voting score", this service allows you to search for consensus microRNA:mRNA predictions.

Total score =