Fig. 4
From: HAPPI-2: a Comprehensive and High-quality Map of Human Annotated and Predicted Protein Interactions
a A comparison of distributions of MetaGene overlap counts for randomized samples (size = 10,000 per sample) of HAPPI-1 and HAPPI-2 database subsets. b A comparison of distributions of MetaGene overlap counts for randomized samples of HAPPI-1 and HAPPI-2 database subsets. The results shown has all randomized samples normalized with the database size differences between HAPPI-1 and HAPPI-2 databases, currently with a size ratio factor of 1:4.85. c A comparison of false positive ratio among HAPPI-2, HAPPI-1 and STRING. Here, the STRING’s confidence score (scale 1000) is scaled to 1 to compare with HAPPI-2’s and HAPPI-1’s confidence score on the x-axis (or reliability index for data from a single source)