Table 3. Agreement between experimental results and simulated outcomes in the presence of abscisic acid (ABA).

Response category	Total number of cases	Simulated interventions for which experimental results exist	CPC range
Equivalent to wild type	22	SCAB1 CA (1b) [83], GEF1/4/10 KO (1a) [40, 105], SPP1 KO (1a) [106]	23.93–24.08
Hypersensitivity	48	SphK1/2 CA [107, 108], HAB1 KO [44, 45], RCARs CA [109], GCR1 KO [110], TCTP CA [32], S1P/PhytoS1P CA [70, 111], ABI1 KO [43, 112], OST1 CA [25], ABI2 KO [43], ERA1 KO [113], ABH1 KO [65], PA CA [114], PLDα CA [114] [36]; NIA1/2 KO (1b) [56], Microtubule Depolymerization CA (2) [115, 116]	24.08–26.7
Close to wild type	25	CPK6 KO [86], cGMP CA [57]; MRP5 KO (1b) [64], InsP3 KO (1b) [59, 117], pHc CA (1a) [78], cGMP KO (1b) [57], V-ATPase KO (1b) [30] [104], NO KO (1b) [56], ROP11 KO (1a) [34], PP2CA KO (1a) [118], NOGC1 KO (1b) [57]	23.92–24.07
Hyposensitivity	22	ROP11 CA [104], GEF1/4/10 CA [40, 105], QUAC1 KO [119, 120], PLC KO [59, 117], PtdIns(4,5)P2 KO [30, 121], PtdInsP4 KO [121], PLDα KO [36], V-PPase KO [30], CIS KO [122, 123], PtdIns(3,5)P2 KO [30], cADPR KO [124], PI3P5K KO [30], GAPC KO [67], ADPRc KO [124, 125]; SLAH3 KO (2) [89]	20.22–23.91
Reduced sensitivity	26	Vacuolar Acidification KO [30], pHc KO [87], ABI2 CA [102], H+ ATPase CA [126], CaIM KO [127], PP2CA CA [118], MPK9/12 KO [29], S1P/PhytoS1P KO [70, 111], SphK1/2 KO [107, 108], PtdInsP3 KO [121], GHR1 KO [63], SLAC1 KO [23], RCN1 KO [128], ROS KO [46], RBOH KO [46], PA KO [114], ARP complex KO [31],AtRAC1 CA [82], SCAB1 KO [83], PLDδ KO [67]; GPA1 KO (2) [87]	5.9–21.02
Insensitivity	15	Microtubule Depolymerization KO [116], K+ Efflux KO [11], KOUT KO [11], Aquaporin (PIP2;1) KO [12], OST1 KO [24, 129], RCARs KO [42, 130], ABI1 CA [102], Ca2+ c KO [122, 123]	0.0–0.02

The interventions considered are a sustained OFF state (denoted KO), corresponding to knockout of a gene, disruption of a process, or subthreshold concentration of a molecule, and a sustained ON state (denoted CA), corresponding to constitutive activity of a protein, activation of a process, or external supply of a molecule. Simulated manipulations of the state of source nodes that do not cause any alterations of the initial condition (e.g., supply of a molecule that is already abundant) can be considered as replicates of the wild-type simulation and are denoted “equivalent to wild type.” Interventions that lead to CPC (see Materials and methods) within the wild-type equivalent range are denoted “close to wild type.” Hyposensitivity refers to reaching closure in all simulations but with a longer time line (more time steps) than the wild-type equivalent simulations, reflected in a CPC smaller than the lowest wild-type equivalent CPC (23.93). Hypersensitivity refers to closure in all simulations with a shorter time line (fewer time steps) than the wild-type equivalent simulations and a CPC larger than the largest wild-type CPC (24.08). Reduced sensitivity means that the percentage of closure stabilizes at a value less than 100%; insensitivity means that the percentage of closure stabilizes at 0%. The node names within each category are grouped according to the results of the comparison with the closest experimental information (reported in the indicated reference) and within each group are shown in increasing order of CPC. Genetic or pharmacological KO experiments are compared with the results of simulated KOs. Experimental overexpression or external supply are compared with the results of simulated CA. We note that overexpression of a protein may not correspond to its CA, especially if its activation involves posttranslational modification. Bold font indicates that the model result is consistent with the experimental data and italic font indicates that the model result is not consistent with experimental data. We categorize the cases of discrepancy between model and experiment into 2 types and indicate the category in parenthesis after the node name. In type 1, the model indicates a response equivalent/close to wild type while the experiments indicate significantly increased (1a) or decreased (1b) response compared to wild type. In type 2, the model indicates significant deviation from wild type (of the kind shown in the first column), and the experiments indicate wild-type response.

Abbreviations: CA, constitutive activity; CPC, cumulative percentage of closure; KO, knockout. All other abbreviations are explained in S3 Table.