Table 5. Natural language descriptions of the formal specifications employed for the rat cardiovascular system dynamics, the uterine contractions of labour, the Xenopus laevis cell cycle, and the acute inflammation of the gut and lung case studies.

MId	SId	Description
1	1	The probability is greater than 0.9 that after initiating the Valsava manoeuvre (time = 5000 ms) the thoracic pressure increases from the baseline value -4 to 16 for 10 seconds (time interval [5001 ms, 14999 ms]), and then drops back to the baseline value -4.
	2	The probability is greater than 0.9 that during the initial phase of the response (time interval [5001 ms, 6500 ms]) the aortic pressure increases and the heart rate decreases.
	3	The probability is less than 0.1 that after the initial response phase (time interval [5001 ms, 6500 ms]) the aortic pressure continues to increase or stay constant, respectively the heart rate continues to decrease or stay constant throughout the remainder of the Valsava interval (time interval [6501 ms, 14999 ms]).
2	4	The probability is greater than 0.9 that the intrauterine pressure increases/decreases with the contractile activity of uterine regions.
	5	The probability is less than 0.1 that the intrauterine pressure decreases when the entire uterus experiences an action potential burst.
	6	The probability is greater than 0.9 that the intrauterine pressure decreases when the entire uterus is in the refractory period.
3	7	The probability is greater than 0.9 that whenever the concentration of CDK1 reaches very high levels (in our case >96% of its maximum value) all cells will divide.
	8	The probability is greater than 0.9 that whenever the average concentration of APC increases and reaches its local maximum value no cell will divide.
	9	The probability is greater than 0.9 that the average concentrations of CDK1, Plk1 and APC increase and then decrease (i.e. oscillate) over time at least three times.
4	10	The probability is greater than 0.9 that if the level of cytoplasm occludin in the lung decreases then eventually the number of ischemic endothelial lung cells will increase.
	11	The probability is greater than 0.9 that always an increase of the cell damage by-product in the gut will lead to an increase of the cell damage by-product in the lung.
	12	The probability is greater than 0.9 that if the level of cell wall occludin in the gut decreases then eventually the amount of solute leaking in the gut lumen will increase.

Each model is identified by an id (column “MId”) and has an associated set of natural language statements. Conversely each natural language statement has a corresponding id (column “SId”) and description (column “Description”).