TABLE 1.

Precision (%) of the top L/5, L/2 and L predicted long‐range contacts on the CASP13 dataset

Model	Short‐range			Medium‐range			Long‐range
	Top‐L/5	Top‐L/2	Top‐L	Top‐L/5	Top‐L/2	Top‐L	Top‐L/5	Top‐L/2	Top‐L
Sequence attention	58.26	41.51	27.95	63.10	45.87	32.51	64.46	52.13	39.82
Regional attention	61.08	41.95	28.38	65.46	48.00	33.65	67.32	54.15	40.96
Combined model	60.94	42.69	28.83	66.45	48.45	34.19	70.73	55.88	42.64
Baseline model	59.00	42.73	28.17	64.29	48.01	33.05	66.31	49.42	36.40