Table 3.

Use expectations to compare the generated models against.

Question to address for model validation	Example used in Leukemia cell lines (Aurich et al., 2015) or NCI-60 cell lines (Aurich et al., unpublished)	Check using MetaboTools*, models, or alternative resource
What metabolic pathways does/does not my model include that would (not) be expected for the target cell?	Transcriptomic data integration caused absence of complex I of the electron transport chain in the models, which complied with literature (Aurich et al., 2015).	model.subsystems, model.rxns, metabolic functions (Thiele et al., 2013)
To what extend does the model capture metabolites detected in the intracellular metabolome?		model.mets
Are the models able to achieve experimental growth rates given the applied constraints?	−Analysis was conducted using biomass constraints (Aurich et al., 2015).	setConstraintsOnBiomassReaction* or changeRxnBounds (add constraint)
	−The vast majority of the models was able to grow at experimental growth rates (Aurich et al., unpublished).	optimizeCbModel (perform FBA)
		(changeObejctive—set objective function in model.c)
Which exchange reactions have been added? Are the cells known to use/ secrete these substrates?	Cancer cells are known to use fatty acids to support their growth (Aurich et al., unpublished).	statisticsAddedExchanges*

The table lists examples of questions and data that can be used to validate contextualized models along with existing examples and MetaboTools functions that can be used for this analysis.