Sciome will showcase its flagship web-based computational toxicology tool OrbiTox at the Society of Environmental Toxicology and Chemistry (SETAC) North America 46th Annual Meeting, taking place November 16-20, 2025, in Portland, OR
OrbiTox is a new interactive tool that’s changing how we understand the safety of chemicals without relying on traditional animal testing. OrbiTox brings together huge amounts of scientific data and uses advanced computer models to predict how chemicals might affect people and the environment.
New approach methodologies (NAMs) are innovative, non-animal testing methods to assess chemical hazards and risks. Among these, in silico tools like read-across and QSAR models use computer algorithms to predict chemical effects based on structure and similarity to other substances. However, one major challenge has been that the relevant data are scattered across many public databases, making it hard for scientists and regulators to see the bigger picture. OrbiTox solves that problem by collecting data from a wide range of sources, including information on nearly 900,000 substances, 22,000 human biological targets, 1,500 biological pathways, and more than 200 test organisms, into one easy-to-use platform. Besides housing multidomain data, OrbiTox contains QSAR models to predict outcome in toxicity assays of mutagenicity, cardiotoxicity, carcinogenicity, and assays conducted under the Tox21 program. Access to multi-domain data, predictive models, and cheminformatics methods in one software tool makes it easier to gain answers to questions like: What are the property profiles of chemicals similar to a specific structure of interest? What are the closest analogues of the chemical of interest, and what is their predicted and experimental toxicity profile? Which chemicals are likely to be active or inactive on a given target of interest for a particular toxicity?
Early findings show that OrbiTox’s QSAR models perform with a high level of accuracy. For instance, models that predict genetic mutations in bacteria using the Ame’s assay, which is the standard test worldwide for determining whether a substance may cause cancer, achieved classification accuracy scores between 0.84 and 0.90. Over 100 QSAR models available in OrbiTox have similar cross-validation performance. Furthermore, what makes OrbiTox especially powerful is its ability to visualize what structural features are responsible for the prediction, helping users understand the potential chemistry behind each result. By uniting advanced computing, large-scale data, and interpretation transparency, OrbiTox gives researchers and regulators a faster, more ethical way to assess chemical safety.
