Computational prediction of RNA-protein interactions

Authors : Hilal KAZAN

Pages : 0-0

Doi:10.21541/apjes.92217

View : 10 | Download : 7

Publication Date : 2016-10-01

Article Type : Research Paper

Abstract :RNA-protein interactions play critical roles in diverse cellular processes including post-transcriptional regulation of gene expression and infection by pathogens. As such, characterization of RNA-protein interactions will lead to a better understanding of these mechanisms and associated diseases.  Experimental methods to determine RNA-protein interactions remain tedious and expensive. An alternative strategy is to use computational methods to predict RNA-protein interactions. Here, we develop a random forest model that uses sequence information of an RNA-protein pair to determine whether they will interact or not. We evaluate our model with three diverse datasets including one dataset that has never been used for this purpose before. For the two other datasets, our model gives a better performance than existing methods. We also show that including features that represent the physico-chemical properties of the protein or RNA secondary structure. Altogether, these results show that RNA-protein interactions can be predicted accurately with computational models. 
Keywords :