Large-scale expression analysis of natural antisense transcripts

点击翻译按钮获取中文摘要

Summary Recent studies have identified a large number of natural antisense transcripts that are transcribed from the genomic loci of well annotated genes, but from the opposite DNA strands. It is still unclear whether the majority of antisense transcripts are functional or merely transcriptional noise. We hypothesized that if antisense transcripts are related to certain physiological functions, they will exhibit variable levels of expression in different types of tissues and the pattern would tend to be evolutionarily conserved. By adapting commercial high-density oligonucleotide microarrays, we developed a cost-efficient approach that can monitor antisense expression across all exonic loci in mammalian genomes. Based on this approach, we will perform systematic profiling of antisense expression in various normal tissues in human, mouse and rat. Coupled with expression analysis of sense transcripts in the same samples, this will define a "double stranded" expression profile at the exon level. The data will be used for comparative analyses to determine if there are conserved patterns of tissue-dependent antisense expression. We will compare the correlations of expression patterns in orthologous pairs of antisense transcripts with the correlation of the expression pattern of permutated pairs. We will also determine whether highly expressed pairs of orthologous antisense transcripts have smaller DNA sequence divergence and whether orthologous pairs of antisense transcripts with smaller DNA sequence divergence tend to have (1) higher correlation in their expression profiles across various tissues, (2) smaller changes in expression level between species, and (3) smaller change in breadth of expression, similar to what have been observed in protein-coding genes. This could either provide substantial evidence for a selective purifying pressure on antisense transcription or favor a "neutral drift" model. The comprehensive datasets will also be used to identify novel antisense transcripts and tissue-specific antisense transcripts for further investigation. Such large-scale analysis of antisense expression will critically evaluate whether antisense transcription is a highly regulated process.