Non-Parametric Change-Point Method for Differential Gene Expression Detection

Abstract

BackgroundWe proposed a non-parametric method, named Non-Parametric Change Point Statistic (NPCPS for short), by using a single equation for detecting differential gene expression (DGE) in microarray data. NPCPS is based on the change point theory to provide effective DGE detecting ability.MethodologyNPCPS used the data distribution of the normal samples as input, and detects DGE in the cancer samples by locating the change point of gene expression profile. An estimate of the change point position generated by NPCPS enables the identification of the samples containing DGE. Monte Carlo simulation and ROC study were applied to examine the detecting accuracy of NPCPS, and the experiment on real microarray data of breast cancer was carried out to compare NPCPS with other methods.ConclusionsSimulation study indicated that NPCPS was more effective for detecting DGE in cancer subset compared with five parametric methods and one non-parametric method. When there were more than 8 cancer samples containing DGE, the type I error of NPCPS was below 0.01. Experiment results showed both good accuracy and reliability of NPCPS. Out of the 30 top genes ranked by using NPCPS, 16 genes were reported as relevant to cancer. Correlations between the detecting result of NPCPS and the compared methods were less than 0.05, while between the other methods the values were from 0.20 to 0.84. This indicates that NPCPS is working on different features and thus provides DGE identification from a distinct perspective comparing with the other mean or median based methods.

Weighted Change-Point Method for Detecting Differential Gene Expression in Breast Cancer Microarray Data

“…According to the ROC curves obtained from simulation study [18], NPCPS was more than 99% correct when for a single gene there are more than 9 samples that contain DGE. However, NPCPS is not very sensitive to the right bound as shown in Fig.…”

“…Besides, as a non-parametric inferential method, NPCPS does not make assumptions about the probability distributions of the variables being assessed, and accordingly, it is not necessary to normalize the microarray data before calculating the test statistic like other parametric methods usually do. By simulation and experiment, NPCPS is effective for DGE detection and outperforms the compared methods with better ROC results in many circumstances [18]. However, the performance of this change-point based method is still limited by the less sensitiveness to the right bound and the statistical significance of the static has not been fully explored.…”

Change-point detection in time-series data by relative density-ratio estimation

“…On the other hand, although retrospective change-point detection requires longer reaction periods, it tends to give more robust and accurate detection. Retrospective change-point detection accommodates various applications that allow certain delays, for example, climate change detection (Reeves et al, 2007), genetic time-series analysis (Wang et al, 2011), signal segmentation (Basseville and Nikiforov, 1993), and intrusion detection in computer networks (Yamanishi et al, 2000). In this paper, we focus on the retrospective change-point detection scenario and propose a novel non-parametric method.…”

FDR control of detected regions by multiscale matched filtering

“…Features of interest can be a set of peaks, a set of change points, or a set of non-zero regions. Peak identification [1][2][3][4][5][6], change point localization [7][8][9][10][11], and region detection [12][13][14][15][16] are common problems in statistics, engineering, medicine, and biology, requiring the development of robust and efficient techniques. Among the features of interest, peaks and change points are zero-dimensional (0D) features while regions are onedimensional (1D) and therefore harder to detect.…”