Fig. 1
From: GECKO is a genetic algorithm to classify and explore high throughput sequencing data
Overview of the GECKO algorithm. Input fastq or bam files from two or more conditions are transformed into a matrix of k-mer counts across all samples. The k-mers for which the counts are below a noise threshold or that do not vary across samples are removed (red dots on the right of the k-mer matrix). The adaptive genetic algorithm randomly selects groups of k-mers from the k-mer matrix to form individuals. These individuals will go through rounds of mutation, crossing-over and selection to discover individuals capable of classifying the input samples with high accuracy
