Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: Revealing and avoiding bias in semantic similarity scores for protein pairs

Figure 2

Semantic similarity scores for pairs of term groups increase as the annotation lengths increase. For each of the 14 semantic similarity measures, the solid circles represent the median score of 10,000 random pairs of term groups annotated with the same number of GO BP terms. The error bars represent the standard variation of the scores (because the similarity scores were positive, we reduced the negative part of error bars). r and p are the Spearman correlation coefficient and the corresponding significance between the median of the scores and the annotation lengths. To make the 14 plots more informative, according to the ranges of the scores (see Table 1), we classify the plots into three groups, in each of which the plots have the same y-axis scale.

Back to article page