Table 1.

Datasets of protein sequences that were used in the development of the tool.

Dataset	Subset	All sequences	Class 0 sequences	Class 1 sequences	Max. length
TemStaPro-Minor-30	Cross-validation	239 629	146 657	92 972	1022
	Testing	41 930	25 580	16 350	1022
TemStaPro-Major-30	Training	943 605	879 799	63 806	9841
	Validation	230 985	215 167	15 818	9387
	Testing	210 137	196 072	14 065	13 394
SAPPHIRE	Testing	742	371	371	1643
iThermo	Testing	562	289	273	3567