Fig. 1

Schematic overview of running an end-to-end computational analysis with the Multioviz platform. a, b The user uploads their own individual-level data or summary statistics derived from an omics study. c Input data are visualized as gene regulatory networks (GRNs). Here, darker node colors denote greater statistical significance for a genomic variable. The mapping within and between molecular levels are given via edges which share the same color as the out-degree node. d Multioviz allows users to visualize and perturb GRNs from a prioritized list of significant molecular variables. e Through the perturbation feature, users can explore their generated GRN and delete nodes, then rerun statistical analyses to produce a new GRN. Subsequent rerunning of the variable selection method regenerates data to be visualized as an updated GRN. f Human-in-the-loop perturbation analyses provide better informed in silico hypotheses to be tested and validated in the wet-lab