Fig. 3
From: Estimating Phred scores of Illumina base calls by logistic regression and sparse modeling
Empirical discrimination powers for three methods: backward deletion with either AIC or BIC, and L 1 regularization. The x-axis is the -log10 (error rate) in the range between 3.3 and 3.7. The y-axis is the empirical discrimination power defined as the largest proportion of bases whose empirical error rate is less than 10−x. L 1 1x, 5x, 50x indicates that the L 1-regularized model is trained with 1-, 5-, 50-folds of 3 million bases, respectively