Epicentre MasterPure RNA isolation Kit, Madison, WI
Label
Cy5
Label protocol
The generated cDNAs from RNA were coupled with Cy5 by dissolving the pellets together with a dried aliquot of dye in 9μl of 0.1M Na-Bicarbonate pH 8.5-9 and incubation in the dark at room temperature for 1.5-3 hours. This reaction was quenched with 4.5μl 4M hydroxylamine pH 8.5-9 and incubation in the dark for 15 minutes.
Epicentre MasterPure RNA isolation Kit, Madison, WI
Label
Cy3
Label protocol
The generated cDNAs from RNA were coupled with Cy3 by dissolving the pellets together with a dried aliquot of dye in 9μl of 0.1M Na-Bicarbonate pH 8.5-9 and incubation in the dark at room temperature for 1.5-3 hours. This reaction was quenched with 4.5μl 4M hydroxylamine pH 8.5-9 and incubation in the dark for 15 minutes.
Hybridization protocol
Unincorporated Cy-dyes were removed with a QiaQuick PCR Purification Kit (Qiagen, Valencia, CA) as follows: 35μl 100mM Na-acetate pH 5.2 and 50 μl water were added to each sample, along with 500μl Buffer PB. Thereafter, the standard QiaQuick protocol was followed, with the exception that 75% EtOH was used in place of Buffer PE. Samples were eluted in water in place of Buffer EB. 0.5-1μg of cDNA from each sample was mixed with ~125μl ArrayHyb microarray hybridization buffer (Sigma-Aldrich) and hybridized to full-genome microarray slides using a GeneTAC HybStation (Genomic Solutions, Ann Arbor, MI)
Scan protocol
images were scanned from the completed hybridization using a GenePix 4000B array scanner (Molecular Devices, Union City, CA).
Description
5 min time point in aerobic-anaerobic shift response. Note: Aerobic standard was not consistently labeled across all time series.
Data processing
Two color cDNA microarray data are never devoid of spurious technical contributions that originate during array printing, as well as during the collection and processing of samples, fluorescent labeling and hybridization and scanning of the microarray images (Balazsi et al. 2003, PNAS). To minimize the effect of such contributions, microarray data were normalized as described before (Tong et al. 2004 BBRC). Briefly, spots were excluded from further analysis if the foreground intensity of less than 50% of the pixels within the spot were above 2 standard deviations of the background. We generated expression data tables in Microsoft Excel, containing the following information for each of the 14,352 entries: block (B), column (X), and row (Y) number, red foreground (f_r) and background (b_r), green foreground (f_g) and background (b_g) intensity. The position of each probe within a block, P is defined by the pair of integers (X,Y). Log ratios were defined as the base 10 logarithm of (F_r-B_r)/(F_g-B_g), where F_r, B_r, F_g and B_g represent the median Cy5 (red) foreground and background, and the median Cy3 (green) foreground and background intensities, respectively. In some cases, when the intensity of the background was higher than or equal to the intensity of the foreground, the resulting log ratios became complex or infinity. These values were eliminated using the find, imag, and isfinite functions in Matlab. Next, data were normalized, by averaging the log ratios resulting from all spots printed by a print tip, and subtracting the resulting average from all the individual log ratios corresponding to the same tip. Finally, all the log ratios of the same gene from each slide were averaged and listed in a new file.
