Eco-CosmePharm · Computational 'eco-toxicity' assessment of pharmaceutical and cosmetics materials, an approach towards a green and sustainable environment

Computational 'eco-toxicity' assessment of pharmaceutical and cosmetics materials, an approach towards a green and sustainable environment

There are varieties of materials utilized or produced in the pharmaceutical industry, but the materials that have a huge impact on the environment are the solvents and the end-products. The pharmaceutical products are released in the environment during manufacturing, usage, and disposal. Some are adsorbed to the soils and get deposited, while most are water-soluble and have low volatility, and thus are transported to water compartments in the environment. The main purpose of any pharmaceutical drug/medicine is to ‘stimulate desirable’ or ‘inhibit undesirable’ physiological responses in humans. However, such potent pharmaceutical products including solvents used, are highly capable to show unforeseen adverse effects to non-target ecological species when released in the environment. Another major concern is that these chemicals unintentionally released in the water, although in low concentrations, might pose a high risk to human health after ingestion of contaminated drinking water over a long period, or even there is an adequate risk of contamination via the food chain. The main objective is to identify and reduce the impact of potentially hazardous pharmaceuticals, cosmetics, and solvents on the aquatic environment. The toxicity-related properties that will be studied include acute and chronic toxicity, biodegradation, and bioaccumulation. The research methodology to perform toxicity assessment will majorly involve Quantitative Structure-Toxicity Relationship (QSTR), which employs several machine learning approaches for understanding the structural features responsible for aquatic toxicity. Conclusion of the actions: We have developed a multi-tasking QSTR model to predict the acute and chronic toxicity of pharmaceuticals and cosmetics. Moreover, we have also developed a multi-tasking QSAR model employing random forest technique to predict the bioaccumulation and a classification-based QSAR model employing linear discriminant analysis technique to predict the biodegradation status of chemicals of our interest. The knowledge gained from the QSAR studies helped us in classifying existing marketed pharmaceuticals and cosmetics into toxic and non-toxic groups. Further, we have also performed the experimental validation of the developed models and the results are encouraging. Interestingly, this project also resulted in the development of highly user-friendly AI-based software tools ‘ProtoML-Basic’ and ‘ProtoML-Mixture’ for efficiently executing several QSAR and machine learning tasks. In the near future, the study will surely help us in screening or designing novel analogs of selected toxic chemicals or to identify alternative chemicals that might show similar desirable physicochemical properties with less or no eco-toxicity.

Текст от CORDIS, на английски · Данни: CORDIS, © Европейски съюз