Figure 8.

SCOP95 predictor vs. baseline 2. The difference between the SCOP95 predictor score and the baseline score (boundaries directly extracted from SCOP templates, as passed as input to the predictor). The size of the blocks represents the error. Although most differences in individual bins are not significant possibly due to the small size of the sample, the overall difference, and difference in the [25,95)% interval are significant, while the gain in the [0,25)% interval is marginal. See text for more details.

Walsh et al. BMC Bioinformatics 2009 10:195   doi:10.1186/1471-2105-10-195
Download authors' original image