One important problem in translational genomics is the identification of reliable and reproducible markers that can be used to\r\ndiscriminate between different classes of a complex disease, such as cancer.The typical small sample setting makes the prediction\r\nof such markers very challenging, and various approaches have been proposed to address this problem. For example, it has been\r\nshown that pathway markers, which aggregate the gene activities in the same pathway, tend to be more robust than gene markers.\r\nFurthermore, the use of gene expression ranking has been demonstrated to be robust to batch effects and that it can lead to more\r\ninterpretable results. In this paper, we propose an enhanced pathway activity inference method that uses gene ranking to predict the\r\npathway activity in a probabilistic manner.Themain focus of this work is on identifying robust pathwaymarkers that can ultimately\r\nlead to robust classifiers with reproducible performance across datasets. Simulation results based onmultiple breast cancer datasets\r\nshow that the proposed inference method identifies better pathway markers that can predict breast cancer metastasis with higher\r\naccuracy. Moreover, the identified pathway markers can lead to better classifiers with more consistent classification performance\r\nacross independent datasets.
Loading....