In BioNERO infers GRNs based on the wisdom of the crowds principle. It evaluates GRNs from different algorithms and calculate average ranks for each interaction and filter the resulting network based on the optimal scale-free topology (SFT) fit.

I would like to know, what SFT cut-off value is used for filtering.

Hi,

There is no cut-off, actually. As the documentation says, the function exp2grn() simply calculates the scale-free topology fit for all subnetworks and selects the one with the best SFT fit (highest R^2).

However, it might happen that:

1. The best network (based on the highest R^2) is not much different from the second best network, and the second best network has much more edges. If more edges is what you want, you might want to pick the second best network, which you can identify from the plot returned by exp2grn(), as in this example.

2. None of the networks have a good SFT fit (say, R^2 < 0.8). Again, that is something you can see in the plot returned by exp2grn(). In that case, you might want to see if a particular method outperforms the combination of methods (the wisdom of the crowds).

Best,

Fabricio