Fig. 5
From: GECKO is a genetic algorithm to classify and explore high throughput sequencing data
GECKO can accurately classify normal and CLL patients using k-mers from bisulfite sequencing data. a GECKO output showing the t-SNE separation of CLL and normal samples using 20 k-mers from the winning individual. b GECKO output of K-mer exploration across 20,000 generations; k-mers that are frequently found in winning organisms are displayed as horizontal lines across generations; dots represent k-mers that were selected in one generation but eliminated in the following generation often due to a decrease in fitness of the model. c IGV screenshots showing the methylation status of normal and CLL samples of regions corresponding to three most frequently used k-mers in winning organisms determined by the Bismark software
