Studying DNA microarray data using independent component analysis

Independent component analysis (ICA) is a statistical technique used to estimate underlying sources from an observed set of data. This work examines the application of ICA on DNA microarray data with the goal of locating distinct, biologically relevant functions from gene expression. Uncovering these functions based on observed gene expression data is shown by selecting outlier values of gene influence from the ICA estimates and examining their corresponding gene annotations. The ICA method is applied to breast cancer data and the analysis shows how the estimated independent components are related to biological functions.