|Application:||Gene expression microarray analysis|
|Number of samples:||9|
|Release date:||Jul 23 2008|
|Last update date:||Feb 18 2018|
|Dataset link||Differential gene expression patterns of the developing and adult mouse cornea compared to the lens and tendon|
Briefly, the corneas were dissected free of the limbus from postnatal day 10 (P10) and seven-week old (adult) mice. To enrich for stromal contribution, the epithelium was scraped off and the cornea placed in TRIzol (Invitrogen Life Technologies, Carlsbad, CA). The scraping off of the epithelium reduced, but did not abrogate epithelial contribution, as judged by the expression of known epithelial markers. The lens and the flexor digitorum longus (FDL) tendon were removed from seven-week old adult mice for isolation of total RNA. For the adult cornea three independent preparations of total RNA was generated. For all the other tissues two independent preparations of total RNA were used. The adult and the P10 microarrays were used as baselines in our previous study of corneal cells in culture to identify genes differentially expressed by cultured cells of the stroma (Chakravarti et al. 2004). The output fluorescence of the scanned images was analyzed first using the Affymetrix Microarray Suite 5.0 software to compute absolute expression, background calculation and detection call (present, absent or marginal). Background signal was between 42 to 55 arbitrary units, while the noise to signal ratio was between 1.42 to 1.78 arbitrary units. The expression data from all nine arrays were further analyzed as follows. The model-based DNA-Chip Analyzer (dChip) software (http://www.dchip.org/) was used to normalize the data from the image files for array-to-array comparison (Li and Wong 2001). This software uses an invariant-set algorithm as a basis for normalization and a model-based expression index to calculate the expression value for each transcript. To identify gene expression patterns that were statistically significantly different between any two given tissue types, we used the following criteria: (1) fold change ≥ 2; (2) difference of mean expression index ≥ 100 arbitrary units; (3) t-statistic, p < 0.05.