Abstract:
|
In many situations, there are different methods for analyzing the same data. For example, several methods exist for finding differentially expressed genes using RNA-seq data. They tend to produce similar, but not identical significant genes and partial rankings of the gene list. Here we consider partial rankings of all genes of a given species. The genes are split into several groups, so that there is a ranking between the groups and not necessarily within each group. A special case is a setup in which the most significant k genes are ranked, with group k+1 consisting of the remaining genes. When comparing different methods applied to the same data, we are interested in how close are their outputs. Prior studies have compared several RNA-seq differential expression methods using empirical results based on real and simulated data. Our approach is based on the statistical distribution of the distance on the set of partial rankings. This allows us not only to measure the relative similarities between different methods, but also to infer whether two methods are significantly different. The outputs of several RNA-seq differential expression methods are analyzed using this approach.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.