Skip to main content
Figure 8 | BMC Bioinformatics

Figure 8

From: A nonparametric model for quality control of database search results in shotgun proteomics

Figure 8

The partitions of k-means clustering before (A) and after (B) normalization (z-score) of the features. Blue and red points represent different clusters. The observations derive from the control dataset. Records with larger Xcorr and ΔCn are more likely to be positive results. The partition given by k-means clustering using the observed values is based on Xcorr; ΔCn has no effect. After normalization, the partition is more consistent with the empirical knowledge.

Back to article page