Skip to main content
Figure 7 | BMC Bioinformatics

Figure 7

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

Figure 7

Illustration of the workflow. The workflow is based on the nonparametric model and the randomized database method. First, the randomized database is constructed and merged with the normal database. Then a database search is performed using SEQUEST. Peptide matches from the randomized database are used to build the mixed Gaussian model. Filter boundaries are determined based on the mixed Gaussian model and the expected FPR, and the normal database matches are filtered. During construction of the nonparametric model, k-means clustering is used to initialize the parameters of the EM algorithm. The red points in the left rectangle are the cluster center on the Xcorr-ΔCn plane. The red pints on the right rectangle denote the matches from the normal database and the blue points are matches from the randomized database.

Back to article page