Adjustment method for microarray data generated using two-cycle RNA labeling protocol
- Equal contributors
1 Center for Quantitative Biology, Peking University, Beijing, 100871, China
2 LMAM,School of Mathematical Sciences, Peking University, Beijing, 100871, China
3 School of Life Science, Peking University, Beijing, 100871, China
4 Center for Statistical Sciences, Peking University, Beijing, 100871, China
BMC Genomics 2013, 14:31 doi:10.1186/1471-2164-14-31Published: 16 January 2013
Additional file 1:
Figure S1. Position of probes on their transcripts (bp) (away from 3’ end) of Affymetrix GeneChip Rice Genome oligonucleotide arrays.
Figure S2. Correlation between position and intensity of probes for present probe sets (By MAS5.0) in Leaf and Leaf Primordium microarray data.
Figure S3. Schematic diagram of Real Time PCR experiments.
Figure S4. The Real Time PCR results for other transcripts show similar trends as in Figure 2.
Figure S5. Estimation of weight for curve adjustment.
Figure S6. Distribution of the Coefficient of Variation (CV) for PM intensities of present probe sets after 3 preprocessing methods.
Figure S7. Hierarchical clustering of 15 microarray samples after 3 preprocessing methods.
Figure S8. Histogram of correlation coefficients between 15 microarray samples after 3 preprocessing methods.
Supplemental Formula. Formula F1: The joint distribution for positions of the new 3 end and 5 end after the 3th shorten A3 and B3: F3(x, y).
Supplemental Results and Discussion. Comparison with Curve Adjustment to demonstrate the necessity of our model for adjusting bias. A simple adjusting method that assigns different weight to probes at different position of transcript according to expression intensity was applied, but the result indicates that direct curve adjustment for microarray data is not suitable and Model adjustment is necessary.
Format: DOC Size: 2.4MB Download file
This file can be viewed with: Microsoft Word Viewer