A microarray analysis of the rice transcriptome and its comparison to Arabidopsis

Abstract

Arabidopsis and rice are the only two model plants whose finished phase genome sequence has been completed. Here we report the construction of an oligomer microarray based on the presently known and predicted gene models in the rice genome. This microarray was used to analyze the transcriptional activity of the gene models in representative rice organ types. Expression of 86% of the 41,754 known and predicted gene models was detected. A significant fraction of these expressed gene models are organized into chromosomal regions, about 100 kb in length, that exhibit a coexpression pattern. Compared with similar genome-wide surveys of the Arabidopsis transcriptome, our results indicate that similar proportions of the two genomes are expressed in their corresponding organ types. A large percentage of the rice gene models that lack significant Arabidopsis homologs are expressed. Furthermore, the expression patterns of rice and Arabidopsis best-matched homologous genes in distinct functional groups indicate dramatic differences in their degree of conservation between the two species. Thus, this initial comparative analysis reveals some basic similarities and differences between the Arabidopsis and rice transcriptomes.

Comparative Network Analysis Reveals That Tissue Specificity and Gene Function Are Important Factors Influencing the Mode of Expression Evolution in Arabidopsis and Rice

“…These results imply that colinear genes between Arabidopsis and rice are, at the regulatory level, not more conserved than genes that have been rearranged since the divergence of both species (Ren et al, 2007). Despite the coexpression of neighboring genes in rice (Ma et al, 2005), currently there is little evidence that general coregulatory mechanisms, complementary to, for example, bidirectional promoters, act on a global scale in plant genomes. Altogether, our findings indicate that orthologs inferred through sequence similarity in many cases do not resemble similar biological gene functions and highlight the importance of incorporating expression information when homologous genes between different species are analyzed.…”

“…Recently, expression data, functional gene annotations, protein-protein interaction data, knockout phenotype information, and cis-regulatory elements have been combined to delineate coexpressed modules, to predict new gene functions, and to identify transcriptional regulatory interactions (Aoki et al, 2007;Vandepoele et al, 2009;De Bodt et al, 2010;Mutwil et al, 2010;Obayashi and Kinoshita, 2010). Although most coexpression approaches have been used to predict different types of gene-gene interactions in model species such as Arabidopsis and rice, interspecies comparisons have identified examples of conserved coexpression modules in plants (Ma et al, 2005;Street et al, 2008).…”

“…Although the genome size of Arabidopsis (Arabidopsis thaliana), a model for dicotyledonous plants, is much smaller than that of rice (115 and 420 Mb, respectively), both species share a large number (56%-77%) of homologous genes (Vandepoele and Van de Peer, 2005;Sterck et al, 2007). Analysis of similarities and differences between the Arabidopsis and rice transcriptomes for similar organ types with custom-made oligomer microarrays revealed that similar portions were expressed in their corresponding organ types (Ma et al, 2005). In addition, evidence was found that a large fraction of rice genes lacking Arabidopsis homologs were expressed (Ma et al, 2005).…”

“…Analysis of similarities and differences between the Arabidopsis and rice transcriptomes for similar organ types with custom-made oligomer microarrays revealed that similar portions were expressed in their corresponding organ types (Ma et al, 2005). In addition, evidence was found that a large fraction of rice genes lacking Arabidopsis homologs were expressed (Ma et al, 2005). Here, we developed a statistical framework to compare the expression of orthologous genes in rice and Arabidopsis.…”

Analysis of Alternatively Spliced Rice Transcripts Using Microarray Data

“…Oligonucleotide array design algorithms are often unable to design oligos specific for all transcripts in a genome, especially for each alternatively spliced transcript. Instead, oligos designed with the sequences shared among alternatively spliced transcripts have been generated to expand the coverage of transcriptomes by whole genome oligo arrays [23,33]. In the NSF45K array, 6,544 oligos are shared by 15,003 gene models, i.e., transcripts.…”