| Summary: | 碩士 === 國立陽明大學 === 生物醫學資訊研究所 === 98 === Identifying genes with monotonic features(i.e., genes with gradually increasing or decreasing expression patterns)over time or over stages of a disease has always been an important issue in microarray data analysis. However, to address this issue classical methods are time-consuming and require experienced bioinformaticians. We have developed a novel method, the Monotonic Feature Selector (MFSelector), based on the nonparametric rank-based Kruskal–Wallis test (H-test), employs a simple approach to analyze data with a view to identifying genes with monotonic features. This new algorithm is also sensitive enough to successfully derive features from heterogeneous gene expression profiles. We have demonstrated the robustness of our algorithm by applying it on two stem cell differentiation data sets and one tumor data set. The MFSelector has successfully revealed monotonicgenes during stem cell differentiation or tumor progression. Some of these genes are known stemness genes (Oct4, Nanog), while many other genes have not been linked to stem cell differentiation or tumorigenesis before. The case studies have helped to get a better understanding of differentiation, stemness, and tumorigenesis. The novel monotonic marker genes discovered from a data set are found to exhibit consistent behavior in another independent data set, demonstrating the utility of the proposed method. The MFSelector R script can be downloaded from: http://microarray.ym.edu.tw/tools/MFSelector/.
|
|---|
