DESCRIPTION (provided by applicant): A novel class of endogenous small RNAs (termed microRNAs, or miRNAs) has recently been discovered that mediates native RNA interference phenomena in mammalian cells. Specific miRNAs are expressed in different cell types, stages of development, and disease states, including certain cancers. Research on miRNA expression patterns requires the development of new analytical methods and tools because the small size of miRNAs (approximately 22 nucleotides) is not readily amenable to existing technologies. Microarrays represent an ideal high throughput method for research on miRNA expression patterns and, in the future, possibly miRNA diagnostics. High throughput expression profiling methods such as microarrays require miRNA samples to be efficiently labeled with a fluorescent dye. The ideal miRNA labeling method would label small RNAs equivalently and without sequence bias, would not alter the melting temperature of the labeled miRNA, would have a sensitive limit of detection, and would require minimal sample manipulation or alteration. No current labeling methods meet all of these criteria. We propose to develop a highly efficient, straightforward RNA labeling technology that would preserve the original RNA sample, add a single label to each RNA target, exhibit no sequence bias in labeling, have a negligible effect on hybridization or melting temperature, and preferentially label small RNA molecules such as miRNA in samples containing complex mixtures of RNA. We will demonstrate the performance of this direct labeling method on a prototype short-oligo miRNA microarray.