Fig. 3: Accuracy of CombFold on Benchmark 3.

a, The top-N (N = 1, 5, 10) success rate of CombFold (blue) and MoLPC (orange). b, Top-1 success rate for homomers and heteromers. c, TM-score comparison for CombFold and MoLPC. d, Predicted confidence versus the TM-score for CombFold. e, The number of complex amino acids versus the top-1 TM-score. f, The success rate of AFM in producing pairwise interactions as measured by the pairwise connectivity versus the TM-score. g, High-quality model of F1-ATPase (top) versus the X-ray structure (bottom). CombFold prediction contains 159 additional amino acids that are not modeled in the X-ray structure, providing full structural coverage. h, Acceptable-quality model of Erwinia ligand-gated ion channel in complex with nanobodies (top) versus X-ray structure (bottom). The channel is accurately modeled; however, the location of nanobodies is incorrect. i, Incorrect model of zinc resistance-associated protein from Salmonella enterica (top) versus X-ray structure (bottom).