Table 1 Comparison of seed sensitivity distributions between models. Seed population sizes for (12,10,0), (14,8,0), (16,6,0) are 352716, 203490 and 54264, respectively. Explanation of columns: y_{min, max}– minimum and maximum sensitivity in the compared model; x _max – maximum sensitivity in the reference model; t_α/2σ _y – half the length of the 95% prediction interval (t_α/2= 1.96 when α = 0.05); T(x, y) = ((a + bx_max + $c{x}_{\max }^2$ + $d{x}_{\max }^3$) - t_α/2σ _y )/y_max, where a, b, c and d are the coefficients of the regression curve, and σ _y is the estimated regression standard error of prediction for a given x value. Because the number of values is large, σ _y ≃ σ for all y. Outliers are determined as those points that satisfy either of the following two criteria: y - $\widehat{y}$ > 2.5σ (U) or $\widehat{y}$ - y > 2.5σ (L). Here, $\widehat{y}$ = a + bx + cx² + dx³.

From: Universal seeds for cDNA-to-genome comparison