Fig. 8.

Parameters estimation workflow. Schematic view of WASABI workflow with 3 main steps: (1) individual gene parameters estimation (red zone), (2) waves sorting (green zone) and (3) network iterative interaction inference (blue zone). Wave concept is introduced in “Results” section. Model parameters (square boxes) are estimated from experimental data (flasks) with a specific method (grey hexagones). All methods are detailed in “Methods” section. Estimated data relative to waves are represented by round boxes. Input arrows represent data required by methods to compute parameters. There are 3 types of experimental data, (i) bulk transcription inhibition kinetic (green flask), (ii) single-cell transcriptomic (blue flask) and (iii) proteomic data (orange flask). Model parameters are specific to each gene, except for θ, which is specific to a pair of regulator/regulated genes. Notations are consistent with Eq. (1), γ_auto represents exponent term of auto-positive feedback interaction. Only d₀(t), d₁(t) and s₁(t) are time dependent. One gene can have several wave times