Reduced DNA methylation in the genome of Arabidopsis suecica. met1-RNAi lines of Arabidopsis suecica Leaf tissue
Extracted molecule
polyA RNA
Extraction protocol
RNA was extracted using TRIzol reagent (Invitrogen Life Technologies), following the manufacture instruction.
Label
Cy3 and Cy5, dye swap
Label protocol
We used 500 ng of mRNA in each labeling reaction using Cy3- or Cy5-dCTP (Amersham Biosciences). The Cy3-dCTP reaction was mixed with the Cy5-dCTP reaction for one hybridization, and then an equal amount of RNA samples was reversely labeled for another hybridization.
RNA was extracted using TRIzol reagent (Invitrogen Life Technologies), following the manufacture instruction.
Label
Cy3 and Cy5, dye swap
Label protocol
We used 500 ng of mRNA in each labeling reaction using Cy3- or Cy5-dCTP (Amersham Biosciences). The Cy3-dCTP reaction was mixed with the Cy5-dCTP reaction for one hybridization, and then an equal amount of RNA samples was reversely labeled for another hybridization.
Hybridization protocol
Each lyophilized probe was re-suspended in 40 μL of hybridization solution (0.25 M Na2HPO4, 0.25 M NaH2PO4, pH 7.4, and 3.5% SDS, w/v). The solution was heated for 2 min at 95 °C, chilled immediately in ice, and applied directly to the array. After covering the array with a 24 × 40 mm coverslip (Sigma, St Louis, MO), the slide was placed in a microarray hybridization chamber (Corning Incorporated, Corning, NY). Hybridization was performed overnight (16 h) at 60 °C in a hybridization oven. After hybridization, the slides were washed for 2 min in 2× SSC, 0.03% (w/v) SDS, 2 min in 0.2× SSC, and 2 min in 0.05× SSC. Immediately after the last wash, the slides were dried by centrifugation (3 min at 500 r.p.m.).
Scan protocol
Slides were scanned using Genepix 4000B
Description
4 replicates with dye-swap experiments
Data processing
The notation Xijkplm is used to denote the mth replicate spot of gene l with feature type p under target condition k labelled with dye j on slide i. After log transformation, Yijkplm = log(Xijkplm).
The anova model for this situation is given by:
Yijkplm = µ + Si + Dj + Tk + Fp + Gl + TFkp + SGil + DGjl + TGkl + FGpl + TFGkpl + ijkplm,where µ represents the overall mean effect, S, D, T, F and G represent main effects from the slide, dye, target (e.g. A.thaliana genome vs. A.arenosa genome), feature type (e.g. oligo vs. gene) and gene, respectively. The interaction terms TF, SG, DG, TG, FG and TFG represent target by feature type, array by gene, dye by gene, target by gene, feature type by gene, and target by feature type by gene interactions, and ijkplm denotes the random error and is used to test for significance of main and interaction effects in the model. Due to confounding and/or aliasing issues involving the slide, dye and target terms, not all two-way interactions are included in the model. The model residuals are assumed to be normally distributed with a common variance (i.e. ijkplm i.i.d. N(0, σ2)). Also, a per gene variance is assumed (i.e. ijkplm independent.
Hypothesis testing
The presence of differential expression in a microarray expression is represented by significant differences in T + TG terms for a particular gene. The following hypotheses are tested to determine whether a gene, g, has undergone differential expression between targets t and t' (e.g. A.thaliana RNA vs. A. arenosa RNA).
H0: Tt + TGtg = Tt' + TGt'g
A standard t-test statistic is used for this comparison, based on the normality assumption for the residuals. To control for multiple testing errors, both Holm's and the false discovery rate (FDR) were employed. Holm's sequential adjustment provides strong control of the family-wise error rate (FWER) below level α with greater power than the standard Bonferroni method (Hochberg and Tamhane, 1987). The false discovery rate (FDR) controlling method of Benjamini and Hochberg (Benjamini and Hochberg, 1995) provides weak control of the FWER, and controls the FDR below level α. The FDR is defined as the expected proportion of incorrect rejections of H0, relative to the total number of rejections. The significance level α= 0.05 was chosen for this study.
