Skip to main content
Fig. 7 | BMC Bioinformatics

Fig. 7

From: Constrained Fourier estimation of short-term time-series gene expression data reduces noise and improves clustering and gene regulatory network predictions

Fig. 7

Post-analysis (k-means clustering) of mouse T cell expression data. A k-means clustering accuracy (Silhouette, see “Methods” section) of de-noised data (red) and raw data (gray) as a function of number of clusters tested from the real data. Difference was statistically significant (\(p<10^{-10}\)). B % improvement of k-mean clustering the de-noised data and the raw data. C The distance to centroids within clusters (calculated by within-cluster sums of point-to-centroid distances, see “Methods” section) of the de-noised data (red) and the raw data (gray) as a function the number of clusters. The de-noised data produced more centered clusters (results significantly different \(p<10^{-10}\). D % improvement of the distance to centroids by using de-noised data

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us