Skip to main content
Figure 4 | BMC Bioinformatics

Figure 4

From: Statistical analysis and significance testing of serial analysis of gene expression data using a Poisson mixture model

Figure 4

Comparison to Bayes error rate for a test of differential expression calculated using a beta binomial model. Using the tag counts from 8 normal brain libraries versus 10 ependymoma libraries, differential expression between these two sample types was assessed using two methods. Plotted are the Bayes error rate described in [7] versus a Poisson mixture model confidence score. For the Bayes error rate (x-axis) smaller values are more significant. The Poisson mixture (y-axis) is a confidence score, so larger values are more significant. Circled are two examples of SAGE tags where one model shows significance while the other does not.

Back to article page