Statistical validation of a comprehensive gene/miRNA expression profile dataset for miRNA:mRNA interaction analysis

MicroRNAs (miRNAs) are small non-coding RNA molecules mediating the translational repression and degradation of target mRNAs in the cell. Mature miRNAs are used as a template by the RNA-induced silencing complex (RISC) to recognize the complementary mRNAs to be regulated. Up to 60% of human genes are putative targets of one or more miRNAs. Several prediction tools are available to suggest the miRNA targets , however, only a small part of them has been validated by experimental approaches. In addition, none of these tools does take into account the network structure of miRNA:mRNA interactions, which we believe is crucial to efficiently predict the miRNA regulation effects in a specific cellular context. We aim to model the miRNA:mRNA interaction network, by including all the miRNAs and mRNAs endogenously expressed in any cellular condition. We started by using as test bed the breast cancer MCF-7 cells. In order to build the miRNA:mRNA interaction model, we collected several miRNA and mRNA expression profiles, by using the Agilent microarray platforms . We analyzed samples derived from the immunoprecipitation (IP) of two RISC proteins, AGO2 and GW182. Specifically, we considered the input, the IP and the flow through samples. We also collected and analyzed miRNAs and mRNAs from polysomal/non polysomal fractions separated through sucrose gradient, as completion of a dataset useful to investigate on miRNA function. The expression level of the top expressed miRNAs has been validated by real time PCR. Due to the peculiarities of our dataset, we used non-standard bioinformatics techniques to preprocess and analyze the obtained expression profiles. As result, we validated the sample extraction techniques (both RISC proteins IP and polysomes isolation), by obtaining expression profile clustering and regression results consistent with the experimental design. Our dataset can then be used to further investigate on miRNA:mRNA interactions, and here we also show our preliminary results in this direction.