Rice Exon Microarray The profiling was conducted with the Rice Exon Microarray (www.ggbio.com/riceexonchip.html;GreenGene Biotech) designed from around 53,000 genes. Among these, 37,400 genes were collected with CAP3 (http://genome.cs.mtu.edu/cap/cap3.html) by clustering 1.1 million ESTs, UniGene Build #60, down-loaded from NCBI (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=unigene) and 17,500 genes were predicted with genezilla prediction program (http://www.genezilla.org/). Around 240,000 exons from these genes were aligned onto 12 rice chromosomes with spidey program (http://www.ncbi.nlm.nih.gov/IEB/Research/Ostell/Spidey/). Selection marker genes such as gfp, gus, hyg, bar, and kan are included. Finally, 320,000 probes were designed by normalizing exons in the way to extract a probe for every 300 bp exon. The average size of probe is 60-nt long to have its Tm value 75 to 85 °C. The microarray was manufactured at NimbleGen inc (http://www.nimblegen.com/). Random GC probes (38,000) to monitor the hybridization efficiency and four corner fiducial controls (225) were included to assist with overlaying the grid on the image. To assess the reproducibility of the microarray analysis, we repeated the experiment two or three times with independently prepared total RNA. The normal distribution of Cy3 intensities was tested by qqline. The data was nomalized and processed with cubic spline normalization using quantiles to adjust signal variations between chips and Rubust Multi-Chip Analysis (RMA) using a median polish algorithm implemented in NimbleScan (Workman et al., 2002. Genome Biol. 3: research0048.1 - research0048.16 ; Irizarry et al., 2003. Nucleic Acids Research 3:e15). Hybridization Genes on microarray