Skip to main content
. 2016 Nov 23;6:37140. doi: 10.1038/srep37140

Figure 1. Workflow of the heuristic model search algorithm.

Figure 1

On the first step the marginal likelihood of the null model is calculated. Then each TF is evaluated as a variable independently and only TFs whose marginal likelihood is higher than the null model’s are further expanded. The highest marginal likelihood of the single TF models is selected as a threshold or bound to evaluate the nested models with two TFs.