Table 2 Effects of including domain-oriented orthology when processing a set of 2000 MAGs. Execution time columns show the total execution time and its increase due to the inclusion of the domain-based pipeline. The increase in execution time is low if we consider the magnitude of the input dataset, whereas the number of predicted orthologs increased by more than one third. OrthoFinder and Broccoli failed to complete the execution for different reasons

From: SonicParanoid2: fast, accurate, and comprehensive orthology inference with machine learning and language models