Fig. 7
From: NeoMutate: an ensemble machine learning framework for the prediction of somatic mutations in cancer
Bar chart of 5-CV GBDT feature importance ranking. The importance of a feature is computed as the (normalized) total reduction of the criterion brought by that feature. The features contributing most to the prediction variable are represented in the left of the plot with highest relative importance scores. The error bar represents the standard deviation across the 5 folds