Figure 6.

Overall distribution of accuracies across all feature set sizes in each phenotype. The format of the plots is the same as in Figure 5. Unlike Figure 5, accuracies were pooled only across the ten feature set sizes and two independent datasets that belong to a phenotype.

Hwang BMC Genomics 2012 13(Suppl 7):S26   doi:10.1186/1471-2164-13-S7-S26