PressPurt: network sensitivity to press perturbations under interaction uncertainty

While the use of networks to understand how complex systems respond perturbations is pervasive across scientific disciplines, uncertainty associated with estimates pairwise interaction strengths (edge weights) remains rarely considered. Mischaracterizations strength can lead qualitatively incorrect predictions regarding system responses as propagate through often counteracting direct and indirect effects. Here, we introduce PressPurt, a computational package for identifying interactions whose must be estimated most accurately in order produce robust network's response press perturbations. The provides methods calculating visualizing these edge-specific sensitivities (tolerances) when one or more edges according variety different error distributions. software requires network represented numerical (quantitative qualitative) Jacobian matrix evaluated at stable equilibrium. PressPurt open source under MIT license available both Python an R hosted https://github.com/dkoslicki/PressPurt on CRAN repository https://CRAN.R-project.org/package=PressPurt.