Incorporating structure to predict microRNA targets.
Proceedings of the National Academy of Sciences of the United States of America. 102:4006-9. Abstract
MicroRNAs (miRNAs) are a recently discovered set of regulatory genes that constitute up to an estimated 1% of the total number of genes in animal genomes, including Caenorhabditis elegans, Drosophila, mouse, and humans [Lagos-Quintana, M., Rauhut, R., Lendeckel, W. & Tuschl, T. (2001) Science 294, 853-858; Lai, E. C., Tomancak, P., Williams, R. W. & Rubin, G.M. (2003) Genome Biol. 4, R42; Lau, N. C., Lim, L. P., Weinstein, E. G. & Bartel, D. P. (2001) Science 294, 858-862; Lee, R. C. & Ambros, V. (2001) Science 294, 862-8644; and Lee, R. C., Feinbaum, R. L. & Ambros, V. (1993) Cell 115, 787-798]. In animals, miRNAs regulate genes by attenuating protein translation through imperfect base pair binding to 3' UTR sequences of target genes. A major challenge in understanding the regulatory role of miRNAs is to accurately predict regulated targets. We have developed an algorithm for predicting targets that does not rely on evolutionary conservation. As one of the features of this algorithm, we incorporate the folded structure of mRNA. By using Drosophila miRNAs as a test case, we have validated our predictions in 10 of 15 genes tested. One of these validated genes is mad as a target for bantam. Furthermore, our computational and experimental data suggest that miRNAs have fewer targets than previously reported.