1. Start
  2. Toolbox
  3. Hazard and exposure screening
A diagram of the tools included in Mistra SafeChem's toolbox

Toolbox: Hazard and exposure screening

Defined approaches are characterised by the combination of in littero, in silico and in vitro hazard prediction methods in a decision framework. The starting point is usually a situation where there is insufficient literature data to say whether a chemical is hazardous in a particular way.

If data is not available, in silico tools are applied as appropriate. The results can be used as the basis for a decision based on certainty, either to advise on hazard or to proceed a biological test to confirm or not the hazardous property. Defined approaches are not necessarily regulatory compliant, based on acceptance guidelines, but allow screening decisions to be made. The current focus is on defined approaches for mutagenicity, endocrine disruption and skin sensitization, where the decision points are clearly defined.

Suspect and non-target screening

An analytical toolbox, including GC/MS (gas chromatography/mass spectrometry) and LC/HRMS (liquid chromatography with high resolution mass spectrometry), that enables broad screening, tentative identification and relative quantification of a priori unknown chemicals, and supports the inclusion of chemical risks in life cycle assessments of chemical processes, such as textile production.
Contact: Ulrika Nilsson, Gunnar Thorsén, Jon Martin

Non-target screening, focus textile

A fully automated solvent-free high-throughput analytical method for the screening of health hazards in textiles. The method is based on automated online thermal desorpion with GC/MS (gas chromatography/mass spectrometry). Under discussion as a standardised method for textile screening.
Contact: Ulrika Nilsson
Solvent-free automated thermal desorption-gas chromatography/mass spectrometry for direct screening of hazardous compounds in consumer textiles External link, opens in new window.
Cellulose Nanocrystals from Postconsumer Cotton and Blended Fabrics: A Study on Their Properties, Chemical Composition, and Process Efficiency External link, opens in new window.
Disperse azo dyes, arylamines and halogenated dinitrobenzene compounds in synthetic garments on the Swedish market External link, opens in new window.
Suspect and non-target screening of chemicals in clothing textiles by reversed-phase liquid chromatography/hybrid quadrupole-Orbitrap mass spectrometry External link, opens in new window.
Manuscript in progress: The fate of hazardous textile pollutants in an upcycling process for post-consumer garments. Contact: Ulrika Nilsson
Ready for use

Non-target screening, surface water monitoring

Procedure for screening and characterising chemicals in water. Water samples are filtered and injected directly into the LC-Orbitrap-HRMS by online solid phase extraction and data acquisition in positive and negative mode with data-independent MS2 analysis. Data pre-processing in MS-DIAL, spectral matching to MassBank and GNPS, and options for molecular networking and in silico structure prediction.
Contact: Jon Martin
Article: Nontarget Analysis of Polluted Surface Waters in Bangladesh Using Open Science Workflows External link, opens in new window.
Status: Ready for use

Non-target sampling of surfaces and air

Pre-cleaned polydimethylsiloxane (PDMS) sheets and foam are exposed to commercial materials (e.g. surface materials in cars) or air (e.g. car cabin) and analysed by target or non-target analysis.
Contact: Jon Martin
Background material: Silicone Foam for Passive Sampling and Nontarget Analysis of Air External link, opens in new window.
Status: Ready for use

AI-based prediction tools

A testbed of AI-based prediction tools has been developed using cutting-edge AI-algorithms. This includes:

  • Endocrine disruption hazard screening suite (consisting of models for twelve endocrine endpoints)
  • CMR hazard screening suite (consisting of models for carcinogenicity, mutagenicity and reproductive toxicity endpoints)
  • ECOTOX hazard screening suite (consisting of models for BCF, bioaccumulation, and persistence endpoints)

Contact: Swapnil Chavan
Article: hERG-toxicity prediction using traditional machine learning and advanced deep learning techniques External link, opens in new window.
Status: Ready for use

Machine learning based tools (1)

The Cytiva machine learning tool in the in silico toolbox includes models for carcinogenicity, mutagenicity, reproductive toxicity, skin sensitization, eye irritation, eye corrosion and 14 endocrine disruption endpoints.
Contact: Ziye Zheng
Article: Manuscript in progress
Status: Work in progress

Machine learning based tools (2)

In silico toolbox with models for 23 endocrine receptors, three ecotoxicological endpoints (biodegradation, bioaccumulation, persistence) and several other toxicological endpoints (carcinogenicity, mutagenicity and reproductive toxicity).
Contact: Ulf Norinder 
In Silico Identification of Potential Thyroid Hormone System Disruptors among Chemicals in Human Serum and Chemicals with a High Exposure Index External link, opens in new window.
Predicting Endocrine Disruption Using Conformal Prediction – A Prioritization Strategy to Identify Hazardous Chemicals with Confidence External link, opens in new window.
Status: Ready for use

Pool of existing computational methods

Assimilation and description of a pool of computational methods for chemical risk and hazard prediction using existing literature data and methods. Includes indication of future development of improved models using state-of-the-art machine learning technologies.
Contact: Swapnil Chavan, Ulf Norinder
Status: Work in progress

Biodegradation prediction

Application of a selection of in silico tools for predicting ready biodegradability and (bio)degradation time/persistence of chemicals and chemical alternatives for substitution. Evaluation of in silico predictions using experimental (bio)degradation tests according to OECD guidelines.
Contact: Gunnar Thorsén, Ioannis Liagkouridis
Status: Work in progress

Biotransformation prediction

Application of a selection of (bio)transformation pathway prediction tools to predict plausible abiotic and biotic transformation pathways and products of chemicals and chemical alternatives. Evaluation of in silico predictions using experimental (bio)degradation tests according to OECD guidelines in combination with suspect screening analytical techniques. A methodology involving the combined use of in silico tools, experimental testing and suspect screening methods for the prediction and identification of transformation products.
Contact: Gunnar Thorsén, Ioannis Liagkouridis
Article: Assessing the Environmental Transformation of Alternative Chemicals Using in Silico Tools, (Bio)Degradation Testing and Suspect Screening – a Case Study of Emerging Alternative Plasticizers External link, opens in new window.
Status: Ready for use

Eye irritation and skin sensitization predictions based on solubility parameters

Prediction of eye irritation and skin sensitization using Hansen solubility parameters and predicted pKa values.
Contact: Martin Andersson, Mikael Kjellin
Article: In Silico Prediction of Eye Irritation Using Hansen Solubility Parameters and Predicted pKa Values External link, opens in new window.
Status: Ready for use

In the in vitro toolbox, Mistra SafeChem has a battery of tools that can be used for defined approaches. These include tools for mutagenic potential using the 2-strain non-GLP micro-Ames assay and skin sensitization using the hCLAT cell assay. It is also possible to perform peptide binding assays for chemical haptenisation using S-9 mixtures. In addition, human hepatocyte toxicity can be assessed using HepG2 as a surrogate for potential liver toxicity. Hormone disruption assays can be performed by a contract laboratory (BDS) using Callux reporter gene constructs.

High-throughput screening methods for combinatorial toxicity testing

High-throughput screening (HTS) methods for combinatorial toxicity testing that can be combined with exposomics and applied in case studies to assess potential human and ecological hazards from mixtures of chemicals.
Contact: Oskar Karlsson
Article: Manuscript in progress
Status: Work in progress

High-content screening protocols for ecotoxicological testing in Daphnia magna

Novel high-content screening (HCS) methods to determine hazard properties and better predict ecotoxicological effects in aquatic keystone species.
Contact: Oskar Karlsson, Magnus Breitholtz
Article: Manuscript in progress
Status: Work in progress

High-content screening of cell phenotypes

Cell painting assay that enables detailed phenotypic cell profiling by using high content screening (HCS), multiplexing of six stains, and automated image analysis. This novel tool captures large amounts of unbiased information to quantify the cellular state following a chemical perturbation.
Contact: Oskar Karlsson
Article: High-content analysis shows synergistic effects of low perfluorooctanoic acid (PFOS) and perfluorooctane sulfonic acid (PFOA) mixture concentrations on human breast epithelial cell carcinogenesis External link, opens in new window.
Status: Ready for use

Protocols for analysing human health effects of textile chemicals

Combination of experimental test systems and mass spectrometry to study skin sensitization and mutagenic effects of textile chemicals and peptide-hapten adducts.
Contact: Ulrika Nilsson, Anders Kallin, Sanja Juric
Manuscript in progress: Health risks from exposure to chemicals in clothing – non-regulated halogenated aromatic compounds. Contact: Ulrika Nilsson.
Status: Work in progress