Provisional Scientific Programme > Terry Speed

Terry Speed

Statistical Analysis of Replicated Microarray Time Series Data

We describe a one-sample multivariate empirical Bayes statistic (the MB statistic) to select differentially expressed genes from replicated microarray time course experiments. We do this by testing the null hypothesis that the expectation of a k-vector of a gene's expression levels is a multiple of 1k, the vector of k 1s. The importance of moderation in this context is explained. Together with the MB statistic we have the one-sample statistic, a variant of the one-sample Hotelling . Both the MB statistic and can be used to rank genes in the order of evidence of nonconstancy, incorporating the correlation structure among time point samples and the replication. In a simulation study we show that the MB statistic and statistic achieve the smallest number of false positives and false negatives, and perform slightly better than the one-sample moderated Hotelling statistic. Several special and limiting cases of the MB statistic are derived, and two-sample versions described. Finally, we illustrate the use of these statistics in two microarray time course studies.

Terry Speed Department of Statistics and Program in Biostatistics University of California at Berkeley.
Speed ’s webpage