Projects


Bio-Data Science

DeepFPlearn

deepFPlearn

AI on the Chemical Universe

We develop AI tools like deepFPlearn to predict the associations between chemicals & effects for 100s of 1000s of compounds in seconds. By including explainable AI and Quantification of Uncertainty, we enhance the credibility of the tool’s predictions for any set of compounds.

more

ChEOS-base

ChEOS-base

Data Integration with Graph Databases

We use a graph database to integrate chemical and toxicity information with monitoring data. It captures relationships between datasets, enabling a comprehensive analysis and providing a thorough understanding of environmental impacts on aquatic ecosystems.

more

ChEOS-mix

ChEOS-mix

Complex Risk Drivers in Water

We use measured chemical concentrations extracted from monitoring data for chemicals measured in European surface waters from the NORMAN network data resource and integrate it with algae, daphnia, and zebrafish toxicity information to evaluate the risk of chemical mixtures on the European scale.

more

AI Training Data Generation

AI Training Data Genaration

Generate transparent data for predictive toxicology

We craft advanced, adjustable pipelines for AI training data generation in predictive toxicology to ensure Reproducibility and future-proofing research. Data extraction, collection, integration, curation, and pre-processing are crucial to providing clean and precise data to machine learning applications.

more


Computational Systems Biology

Multi-omics data integration

multi-omics

Shaping multi-omics data for risk assessment

We are pioneering projects in multi-omics data integration, shaping their applicability for systems toxicology and risk assessment including best practices for study design and multi-omics data analysis. We are developing cutting-edge multi-omics pathway enrichment methods that embrace the complexity of biological networks and foster regulatory decision-making through pathway-based dose-response modeling.

more

Omics-based grouping of chemicals

EFSA_CAGs

Inference of chemical grouping from processed OMICS data in CTD

The EFSA Tender aims to harness omics data from the CTDbase to predict chemical groupings and enhance understanding of their effects on human health and the environment. We seek to align these groupings with established CAGs, potentially facilitating the identification of additional chemicals, beyond the realm of pesticides , that might be included in these CAGs

more

XomeTox

XomeTox

Evaluating multi-omics integration for assessing rodent thyroid toxicity

The CEFIC LRI project XomeTox pioneers the use of multi-omics data in a novel, dynamic approach to unravel the complex molecular effects of toxicants, aiming to redefine our understanding of toxicity pathways. By integrating diverse omics datasets in a time- and concentration-resolved study, XomeTox seeks to discern adaptive from adverse molecular responses, paving the way for novel insights into regulatory network perturbations and toxicity prediction.

more


Toxicokinetic Modeling

Permeability prediction

Membrane permeability

Membrane permeability predicted by the solubility-diffusion model

Using our own experiments and literature data, we show that the intrinsic membrane permeability of chemicals is well-predicted by the solubility-diffusion model, not only for artificial phospholipid membranes, but also biological Caco-2 and MDCK cell membranes.

more

Superhydrophobic chemicals

Bioaccumulation of superhydrophobic chemicals

Bioaccumulation potential of super hydrophobic chemicals

The assessment of the bioaccumulation of highly hydrophobic chemicals faces challenges, including predominant exposure through food, difficulties in conducting standard experiments, and unreliable estimation methods. Here, we aim to overcome these issues in evaluating superhydrophobic substances.

more

Predicting uncoupling toxicity

Protonophoric uncoupling

From molecular structure to uncoupling toxicity

We developed a mechanistic model to predict the toxicity of protonophoric uncoupling based on the molecule's chemical structure. 

more

Active transport

Active transport

Role of active efflux in membrane permeability    

We use a combined approach of mathematical modeling and transwell experiments to characterize the facilitated transport of chemicals through cellular membranes by efflux pumps.

more

Partition coefficients

Partitioning

Partition coefficient prediction for neutral and ionic compounds

Our group’s expertise extends to the experimental determination and in silico prediction of partition coefficients; see the LSER database for neutral chemicals (http://www.ufz.de/lserd). For the prediction of ionic membrane/water partition coefficients, we recommend the commercial software COSMOmic.

more

Bioaccumulation

In vitro in vivo extrapolation

In vitro – in vivo extrapolation

We developed a comprehensive, easy-to-handle model for fish and rat that extrapolates in vitro biotransformation data and uses the extrapolation results for bioaccumulation prediction.

more