Frederic-Bois

Contact

Alternatives to animal experiments

Replacement of animal experiments with combination of innovative mathematical modeling and in vitro assays is a major research application of the chair's research activity.

Research projects:

• • EU-ToxRisk. Flagship project for the development of in vitro toxicology.

  • Euromix. Will advance the state of science for predicting effects of mixtures of food contaminants.

  • StemBANCC. An extensive effort to develop induced-pluripotent human stem cells for therapeutic and predictive toxicology applications.

  • SEURAT/COSMOS. We developed generic PBPK models including dermal absorption and applied then to ab initio predictions of the pharmacokinetics of cosmetic ingredients in humans.

  • PREDICT-IV. We modeled the nrf2 pathway in kidney and liver cells. Coupled with PBPK modeling this allowed us to better predict repeated dose toxicity from in vitro experiments. Extensive in vitro PK modeling was also performed in a Bayesian statistical framework.

  • DYSHORMO. How to extrapolate the results of alternative toxicity testing to in vivo conditions? For this, we developed a computational model of the ovarian steroidogenesis network in order to quantify and predict the endocrine disruption following chemicals exposure.

  • ToxOnChip. We developed, based on in vitro experiments, a computational model of the blood-testicular barriers' integrity.

Predictive toxicology

The European Project PREDICT-IV aims to develop the coupling of in vitro and in silico methods for the prediction of systemic repeated dose toxicity. Pr. Bois is responsible for the modeling work-package (WP 5)

Pharmacokinetics

Our major focus in this area is on physiologically based pharmacokinetic (PBPK) modeling. PBPK modeling is coming of age and is the object of routine applications in drug development and health risk assessments, at least in the USA. There is still some work to do for its regulatory acceptance in Europe and other places, and its exact role and deployment in quantitative in vitro to in vivo extrapolation remains to be fully articulated (see Adler et al., 2011, Arch. Toxicol., 85:367 Bioinformatics -485, doi:10.1007/s00204-011-0693-2). We try to help in that debate.

Dr. Cheng has recently completed two analyses of omics data obtained in microfluidic biochips (Cheng S., Prot J.-M., Leclerc E., Bois F., Zonation-related pathways in human hepatocellular carcinoma cells in dynamic vs. static culture conditions, BMC Genomics, submitted; another article is in preparation).

We also used Bioconductor in order to analyze microarray data for the project ToxOnChip - Modeling.

We are also working, in collaboration with UTC LMAC lab, on Bayesian inference on graphs (with application to network structure inference) (project Prior-Motives).

A more advanced research frontier is the extension of PBPK modeling with systems biology models of metabolism and pharmacology or toxicity pathways. We are actively researching those connections (see for example Cheng and Bois, 2011, Environ. Health Perspect., http://dx.doi.org/10.1289/ehp.1103510).

Another topic of our research on pharmacokinetics is population variability. That is very much linked to multilevel statistical models. We have developed a Bayesian approach to such inference and simulation (for a review : Bois et al., 2010, Toxicology, 278:256-267, doi:10.1016/j.tox.2010.06.007).

Risk assessment

• • Reviews for the US National Academy of Sciences as member of US National Research Council Standing Committee on Risk Analysis Issues and Reviews.

  • Applications of our work to risk assessment tools and concepts were developed in the recent projects 2-FUN, INTARESE, and NexGen.

  • Frédéric Bois is also a member of the French Committee for Prevention and Precaution.

Recent publications

• • ORCID: http://orcid.org/0000-0002-4154-0391

  • Profile on Research Gate

  • Profile on Google Scholar