Fig. 2
From: iMOKA: k-mer based software to analyze large collections of sequencing data
iMOKA accurately predicts breast cancer subtypes. a The features output by all benchmarked approaches are evaluated for their capacity to classify breast cancer subtypes using Random forest’s oob score plotted as a function of the number of the best features output by each approach. b Screenshot of the iMOKA output with each k-mer sequence, their rank in the classification of breast cancer subtypes, and where these sequences map to on the genome. c Screenshot of the iMOKA display showing k-mer counts of the 3 highest ranking k-mers across the 4 subtypes. d Gene ontology of the genes overlapping the k-mers selected by iMOKA