The Importance of Simple Tests in High-Throughput Experiments: An Adventure with cDNA Microarrays Keith Baggerly M.D. Anderson Cancer Center Abstract: Microarray expression profiles can provide quick summaries of the relative mRNA levels for thousands of genes at once, and can consequently allow us to ask very sophisticated biological questions. However, careful processing and manipulation of the data is essential, because our intuition about what the answers ``should look like'' is very poor when we have to juggle thousands of things at once. Consequently, simple tests are very important. Using a case study done on mice (Pritchard et al, PNAS 2001), we describe how one such test questioning the validity of the assumption that "most genes don't change" led to a better method of normalization for the data set at hand, and how this in turn led to the revelation of a more severe problem in data organization and sample mixups. We then describe how we were able to fix this problem by devising a new test and coupling it with the known biology.