Table 1. Thermodynamic analysis of the d3'EBS1*/IBS1* equilibrium by different methods.

	Imaging condition	Fraction of docked d3'EBS1* (%)	K a (L µmol−1)
Gaussian fitting (method 1)	no M(II)(NO3)2	8.2±3.0	3.6±1.3
	1 mM Ni(NO3)2	18.5±6.1	9.1±3.7
	1 mM Co(NO3)2	28.6±5.9	16.1±4.6
Dwell time analysis (HMM, method 2)	no M(II)(NO3)2	15.6±6.4	7.4±3.6
	1 mM Ni(NO3)2	24.6±6.3	13.1±4.5
	1 mM Co(NO3)2	33.3±8.4	20.1±7.5
Histogram thresholding (method 3)	no M(II)(NO3)2	7.8±2.6	3.4±1.2
	1 mM Ni(NO3)2	16.3±5.4	7.8±3.1
	1 mM Co(NO3)2	25.5±6.0	13.7±4.5

The experimental error was estimated by bootstrapping and accounts for 99.7% of the variability observed (3σ _boba, “68–95–99.7 rule” [66]). Association constants K _a were calculated from normalized cumulated FRET histograms or dwell times under the assumption that [IBS] = [IBS1*]_tot as described in the Supplementary Information S1 (Eqs. S1 and S2).