In a nutshell

CBaSE, the method described in Weghorn & Sunyaev (2017), is built around a hierarchical model framework, which estimates gene-specific background substitution rates. It does so based solely on the input somatic mutation data, without the need for knowledge of external mutation rate covariates. Using the substitution rate estimates together with a nucleotide context-dependent mutation matrix, excess or deficit of nonsynonymous substitutions can be compared to the expectation under neutral evolution. The results are gene-specific estimates of positive and negative selection.

Please cite