Table A5.

The Power and SiGMA models give the best fit if we fit the models to subsets of data experiment-wise. As we described in Materials and methods, each Datasets 1–4 has three experiments performed in duplicates. If we divide the data based on the three Experiments 1, 2 and 3 then the Power model gives the best fit for Experiment 1 and 3. For Experiment 2, the SiGMA model gives the best fit. The description of the table remain same as that of Table 1.

Experiment 1  Dataset   $\mathit{n}=125$
Models	$\alpha$	r	k	h	n	$g_0$	$g_1$	$g_2$	SSR	AIC	$\Delta$	w
MA	2.57	0.65	3.84 $\times {10}^{-7}$						34	201	27	0
Sat	2.44	0.67	4.75	2.26 $\times {10}^6$					27.8	177	3	0.18
Power	2.39	0.67	0.003		0.44				27.15	174	0	0.8
SiGMA	2.43		3.22 $\times {10}^{-7}$			9.6 $\times {10}^{-8}$	0.7	12726	28.4	182	8	0.015
Experiment 2  dataset   $n=126$
Models	$\alpha$	r	k	h	n	$g_0$	$g_1$	$g_2$	SSR	AIC	$\Delta$	w
MA	3.47	0.84	3.84 $\times {10}^{-7}$						32	191	29.2	0
Sat	3.32	0.86	4.8	2.78 $\times {10}^6$					27	174	12.2	0.002
Power	3.2	0.86	0.005		0.42				25	164	2.2	0.25
SiGMA	3.18		3.07 $\times {10}^{-7}$			0.018	0.84	6448	24.2	161.8	0	0.75
Experiment 3  dataset   $n=120$
Models	$\alpha$	r	k	h	n	$g_0$	$g_1$	$g_2$	SSR	AIC	$\Delta$	w
MA	2.69	0.67	3.84 $\times {10}^{-7}$						18	121	61.6	0
Sat	2.55	0.7	4.8	2.45 $\times {10}^6$					12.5	79.5	20.1	0
Power	2.47	0.7	0.005		0.41				10.6	59.4	0	0.86
SiGMA	2.50		3.22 $\times {10}^{-7}$			1.08 $\times {10}^{-7}$	0.72	8650	10.76	63	3.6	0.14