saezlab / corneto

Unified knowledge-driven network inference from omics data
https://saezlab.github.io/corneto/
GNU General Public License v3.0
26 stars 3 forks source link

Add inhibitory/activatory edges to unknown perturbations in inverse carnival #5

Open pablormier opened 2 years ago

pablormier commented 2 years ago

If perturbations have a weight of 0, add both edges (activatory/inhibitory) to each perturbation so the optimization process takes one. In the current version (0.9.0-alpha.6) if the weights are 0, only activatory edges are added, biasing the results.