Skip to main content
Figure 6 | BMC Bioinformatics

Figure 6

From: On the analysis of glycomics mass spectrometry data via the regularized area under the ROC curve

Figure 6

Three m/z value area to distinguish cancer from non-cancer. Plot of m/z values areas which can discriminate ovarian cancer (red) from non-cancer (black). The upper left plot is for m/z value 712.28; the upper right plot is for m/z 915.38 and lower plot is for m/z 1442.72. The m/z values are visually larger for cancer patients. For quantification purpose, simple descriptive statistics for the three areas are reported as the following: for m/z value 712.28, mean(standard deviation) of non-cancer samples are 4.506(3.181) and cancer samples are 10.633(3.615). The FDR-adjusted p-value for the comparison is 7.827e-07; for m/z value 915.38, mean(standard deviation) of non-cancer samples are 3.667(2.614) comparing to cancer samples 8.443(3.307). This results in FDR-adjusted p-value of 7.642e-06. The higher m/z 1442.72 has 2.958(2.435) for non-cancer samples of mean(standard deviation), while cancer samples are 7.275(3.1). The FDR-adjusted p-value for high mass area is 1.841e-05.

Back to article page