Fig. 3From: ISOWN: accurate somatic mutation identification in the absence of normal tissue controlsISOWN validation using different machine learning algorithms for six whole-exome sequencing datasets. NBC (green), LADTree (red), and random forest (blue) were trained based on a gradually increasing number of samples (x-axis). The F1-measure was calculated based on a held-out independent sample set across six cancer datasetsBack to article page