Fig. 4.

Performance of CoreCruncher relative to the size of the data set. The core genomes of different sets of genomes of E. coli were built using identical parameters and the same pivot genome (minimum protein identity threshold of 95% and a minimum genome frequency of 90%, nonstringent option). The core genomes were built for four data sets composed of 10, 100, 1,000, and 10,000 genomes of E. coli randomly sampled using the same desk computer (Mac Pro) for each CoreCruncher run. (A) Runtime of the core genomes across the four data sets (note that the x-axis and the y-axis are both in log-scale). (B) Size of the core genome obtained for the four data sets.