Figure 2.

Comparison of p-values for the new and old finite-size corrections using the BLOSUM62 scoring matrix and 11 + g affine gap penalty for equal sequence lengths (m=n) 40, 100, 200, and 400. Figure 2 plots <a onClick="popup('http://www.biomedcentral.com/1756-0500/5/286/mathml/M17','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1756-0500/5/286/mathml/M17">View MathML</a> against p, where <a onClick="popup('http://www.biomedcentral.com/1756-0500/5/286/mathml/M18','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1756-0500/5/286/mathml/M18">View MathML</a> is the calculated p-value and p is the p-value from the simulation. Thus, a perfect p-value estimate corresponds to the horizontal line y = 0. The red line shows the results from the new finite-size correction; the black line, the old finite-size correction.

Park et al. BMC Research Notes 2012 5:286   doi:10.1186/1756-0500-5-286
Download authors' original image