Skip to main content
Figure 4 | BMC Systems Biology

Figure 4

From: Quantitative maps of genetic interactions in yeast - Comparative evaluation and integrative analysis

Figure 4

Pairwise predictive accuracies between the datasets. The prediction accuracy was assessed using the area under the ROC curve (AUC). when predicting the pairs within 3% of the largest interactions levels in the datasets. (A) SGA - GIM data pair, (B) SGA - E-MAP data pair, and (C) E-MAP - GIM data pair. The AUC value was calculated for different versions of the two datasets: Fitness matrix, original double-mutant fitness measurement; SGA custom score, interaction score provided in the SGA dataset; GIM/E-MAP median score, the median estimate for the single-mutant effects with product scoring function in the GIM/E-MAP data. QMA fixed/adjusted, matrix approximation-based score with its two pre-defined settings for scoring positive interactions (see Additional file 1).

Back to article page