Additional file 1: Figure S1. of Differentially expressed genes during the imbibition of dormant and after-ripened seeds â a reverse genetics approach

Microarray quality and reproducibility. All 28 ATH1 arrays used showed after hybridization similar patterns of intensity (A and B). Slide hybridization patterns were inspected manually without detecting artefacts. The RNAs used as templates for cRNA synthesis were shown to be intact based on Bio-analyzer 2001 analysis of both RNA template and biotinylated cRNA. In agreement with this were the hybridization patterns of control genes on the slide showing a near-identical pattern of hybridization (c). The uniformity of normalized unscaled SE (NUSE) and relative log expression (RLE) indicate high quality and uniformity of the hybridization data (d and e) (1). Raw intensity data were subjected to RMA normalization (2), which kept the uniformity of general levels between the different slides (f and g). Between replicate reproducibility of the experiment was high, exemplified by the high correlation between the data of two biological replicates (h). Array 1 and array 2 are hybridized with cRNA from different replicates of Ler seeds. Figure S2. Spatial and temporal expression patterns of the selected dormancy and after-ripening up-regulated genes. Mean relative expression of (a) D-up and (b) AR-up genes across the Arabidopsis germination time course in the micropylar and chalazal endosperm (MCE) and radicle and hypocotyl (RAD) in dry, 1, 3, 7, 12, 16, 20, 25, 31 and 38 hours after imbibition. Data was taken from Seed EFP Browser ( http://www.bioinformatics.nl/efp/cgi-bin/efpWeb.cgi ). Figure S3. Log2 expression differences for the D-up and AR-up genes that are presented in Fig. 1c and d. (a) Heat map showing Log2 expression differences of the 245 NILDOG1 D-up genes (P < 0.0001) in NILDOG1 and the other genotypes. (b) Log2 expression differences of the 159 NILDOG1 AR-up genes (P < 0.0001) in NILDOG1 and the other genotypes. (XLS 81 kb)