view article

Figure 2
Cumulative normalized metrics (the higher the better) for each of the benchmarked methods for our five test sets. Each metric is normalized by the maximum value over the five test sets, and the decreased metrics are inverted to have better values close to 1. Challenge-best means the best solutions from the RNA-Puzzles and CASP-RNA competitions (and corresponds to different solutions for each challenge). The types of methods are also mentioned with the abbreviations DL for deep learning, TP for template-based and Ab for ab initio. Methods are sorted by release time (except for challenge-best). AlphaFold3 (Context) represents the predictions of AlphaFold3 for 113 structures of the RNA3DB_0 data set with the context of the structures added as input.

Journal logoSTRUCTURAL
BIOLOGY
ISSN: 2059-7983
Follow Acta Cryst. D
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds