|
Using expression profiling data to identify human microRNA targets
Jim C. Huang1,6, Tomas Babak2,6, Timothy W. Corson2,3, Gordon Chua4, Sofia Khan2,3, Brenda L. Gallie2,3, Timothy R. Hughes2,4, Benjamin J. Blencowe2,4, Brendan J. Frey1,4,5,7 and Quaid D. Morris2,4,5,7 1 Department of Electrical and Computer Engineering, University of Toronto, 10 King's College Rd., Toronto, ON, M5S 3G4, Canada 2 Department of Molecular and Medical Genetics, University of Toronto, 1 King's College Rd., Toronto, ON, M5S 1A8, Canada 3 Division of Applied Oncology, Ontario Cancer Institute/Princess Margaret Hospital, University Health Network, Toronto, ON, M5G 2M9, Canada 4 Banting and Best Department of Medical Research, University of Toronto, Toronto, ON, 160 College St., M5G 1L6, Canada 5 Department of Computer Science, University of Toronto, 10 King's College Rd., M5S 3G4, Canada 6 These authors contributed to this work equally 7 To whom correspondence should be addressed: frey@psi.toronto.edu, quaid.morris@utoronto.ca |
We demonstrate that paired expression profiles of microRNAs and mRNAs can be used to identify functional microRNA-target relationships with high precision. We used a Bayesian data analysis algorithm, GenMiR++, to identify a network of 1,597 high-confidence target predictions for 104 human microRNAs supported by RNA expression across 88 tissues and cell types, sequence complementarity and comparative genomics. We experimentally verified our predictions by investigating the result of let-7b down-regulation in retinoblastoma using quantitative RT-PCR and microarray profiling: some of our verified let-7b targets include CDC25A and BCL7A. Compared to sequence-based target predictions with low GenMiR++ scores, those supported by a high GenMiR++ score have more consistent Gene Ontology annotations and have a much greater enrichment for mRNAs expressed in let-7b-depleted retinoblastoma and down-regulated in response to let-7b transfection.