Industry: Life Sciences and Healthcare
Business Challenge: Chemists and chemical engineers can unknowingly design consumer products with inherent safety problems. This is due to the lack of reference data and analytical tools needed to model and characterize the broad range of hazards within food, packaging materials, cosmetics, pesticides, and more recently, nanomaterials. For improved consumer safety and a healthier planet, we need to shift away from our reliance on downstream in vivo regulatory testing to continuous in vitro/in silico toxicity screening during all phases of chemical product development.
IOMICS Corporation utilizes multiple cheminformatics techniques and adaptive deep learning to develop Quantitative Structure-Toxicity Relationship (QSTR) decision models. Its QSTR models have demonstrated efficacy in identifying structural patterns that are inherently toxic or that allow hazards to form later from heat, chemical exposure, or other stressors. IOMICS’ innovative use of machine learning and green-design principles has permitted effective simulation of toxicity potential for a diverse range of compounds, species, and endpoints.
AI-enhanced QSTR models represent a new class of interactive analytics that will enable the chemical industry to explore alternative materials for a broad range of consumer products and do so earlier in the product design cycle. This pilot project demonstrated that early and continuous in silico safety screening is a viable strategy that, when used in concert with in vitro testing, increases information yield while reducing R&D costs and downstream product liability. In collaboration with corporate partners, IOMICS has moved the industry closer to an era of rapid, low cost, animal-free safety screening for a broad range of consumer products.