Cellular transcriptional profiling in influenza A virus-infected lung epithelial cells: The role of the nonstructural NS1 protein in the evasion of the host innate defense and its potential contribution to pandemic influenza -- Data Supplement - Supporting Information -- Proceedings of the National Academy of Sciences

Supporting information for Geiss et al. (2002) Proc. Natl. Acad. Sci. USA, 10.1073/pnas.112338099

Supporting Text
Microarray Conditions and Image Analysis.
Two sets of microarrays were used in this analysis containing overlapping gene sets. Set 1 (arrayed at "low" density) consisted of three arrays each with 4,224 individual clones arrayed in duplicate and a control set (384 genes) for a total of 13,056 clones (which corresponds to approximately 12,130 unique IMAGE cDNA clones). Set 2 (arrayed at "high" density) consisted of two arrays each containing 7,296 clones in duplicate and the control set for a total of 14,976 individual clones on two separate arrays (approximately 13,597 unique IMAGE cDNA clones). Every experiment involved the use of two slides in which the dye labels had been reversed for a total of at least four measurements/gene/experiment (see ref. 1 for details). Biological variability was assessed by repeated microarray assays from independent infections for each virus type [three times each for delNS1 virus-infected cells and twice each for wt and NS1(1-126) virus-infected cells]. Detailed protocols can be found at http://ra.microslu.washington.edu/Website/archive_protocol.html.

Microarrays were scanned at two wavelengths with an Avalanche II scanner (Molecular Dynamics). The resulting images were quantitated using a custom spot-finding program, spot-on image (1). Raw data and sample information were entered into a custom designed database, expression array manager, which automatically uploads quantitation data and two-color images into Rosetta Biosoftware's resolver version 3.0 (Rosetta Biosoftware, Kirkland, WA). This software is a high-end package for the storage and analysis of microarray expression data. It is integrated with an oracle database and a java-based user interface. It implements most of the statistical procedures (clustering, trend analysis, etc.) that are commonly used to analyze microarray data. It also implements a sophisticated error model to compensate for biological and experimental variation.

1. Geiss, G. K., An, M. C., Bumgarner, R. E., Hammersmark, E., Cunningham, D. & Katze, M. G. (2001) J. Virol. 75, 4321-4331.