|
Abstract:
|
There have been discussions about detecting differentially expressed (DE) genes in some but not all samples in a disease group for a two-class microarray. This has become increasingly useful in cancer studies, where oncogenes are activated in only a minority of samples. I propose a new statistic called outlier deviation-sum (ODS) to detect DE genes in two-group cancer expression datasets. In simulated examples, I compare ODS to the t-statistic, the outlier robust t-statistic by Wu and the fold-change by Tusher. I find that ODS performs consistently as the best method in terms of accuracy and reproducibility of its gene list under a wide range of conditions. In a real example, ODS provides useful findings beyond those of t-statistic, outlier robust t-statistic and fold-change.
|
