Science Inventory

In Silico Dynamics: computer simulation in a Virtual Embryo (SOT)


Knudsen, T. In Silico Dynamics: computer simulation in a Virtual Embryo (SOT). Presented at SOT National Meeting, Baltimore, MD, March 12 - 16, 2017.


Presentation at SOT annual meeting session on New strategies to drive quantitative systems toxicology for chemical safety assessment.


Abstract: Utilizing cell biological information to predict higher order biological processes is a significant challenge in predictive toxicology. This is especially true for highly dynamical systems such as the embryo where morphogenesis, growth and differentiation require precisely orchestrated interactions between diverse cell populations. In patterning the embryo, genetic signals setup spatial information that cells then translate into a coordinated biological response. This can be modeled as ‘biowiring diagrams’ representing genetic signals and responses. Because the hallmark of multicellular organization resides in the ability of cells to interact with one another via well-conserved signaling pathways, multiscale computational (in silico) models that enable these interactions provide a platform to translate cellular-molecular lesions perturbations into higher order predictions. Just as ‘the Cell’ is the fundamental unit of biology so too should it be the computational unit (‘Agent’) for modeling embryogenesis. As such, we constructed multicellular agent-based models (ABM) with ‘CompuCell3D’ ( to simulate kinematics of complex cell signaling networks and enable critical tissue events for use in predictive toxicology. Seeding the ABMs with HTS/HCS data from ToxCast demonstrated the potential to predict, quantitatively, the higher order impacts of chemical disruption at the cellular or biochemical level. This is demonstrated by specific AOPs that integrate quantitative information for systems such as VEGF-mediated angiogenesis (angiodysplasia), androgen-mediated urethral closure (hypospadias) and TGFb-mediated tissue fusion (cleft palate). Computer models that virtually integrate from real (in vitro) or synthetic (in silico) data provide a platform to translate biomolecular lesions into higher-order emergent responses for predictive toxicology. (Disclaimer: this abstract does not reflect EPA policy).

URLs/Downloads:   Exit

SOT_AOP43_TUES_FINAL_M.PDF   (PDF,NA pp, 3698.244 KB,  about PDF)

Record Details:

Product Published Date: 03/16/2017
Record Last Revised: 08/31/2017
OMB Category: Other
Record ID: 337434