Evaluating different web applications to assess the toxicity of plasticizers

Area of Science:

• Environmental Chemistry
• Toxicology
• Computational Chemistry

Background:

• Plasticizers enhance polymer flexibility but pose environmental and health risks due to leaching.
• Concerns exist regarding water and soil pollution and direct human health impacts from plasticizers.

Purpose of the Study:

• To evaluate web applications for predicting the toxicological properties of various plasticizers.
• To identify specific plasticizers with mutagenic, carcinogenic, or ecotoxicological potential.

Main Methods:

• Utilized web-based tools including BOILED-Egg, LAZAR, PROTOX-II, CarcinoPred-EL, and VEGA.
• Employed predictive models such as Random Forest (RF), XGBoost, KNN/IRFMN, and Sarpy/IRFMN for toxicological assessments.
• Assessed endpoints including mutagenicity, carcinogenicity, bee toxicity, and acute fish toxicity (LC50).

Main Results:

• Bisphenol F was predicted as mutagenic by LAZAR.
• Several plasticizers (BBP, DBP, DEHP, DNOP) were predicted as carcinogenic using CarcinoPred-EL models.
• Multiple phthalates (BPF, di-n-propyl phthalate, diallyl phthalate, dibutyl phthalate, diisohexyl phthalate) showed strong toxicity to bees.
• 19 plasticizers exhibited high acute toxicity to fish (LC50 < 1 mg/L).

Conclusions:

• Web-based tools effectively predict plasticizer toxicity across various endpoints.
• Identified specific plasticizers of concern for mutagenicity, carcinogenicity, and ecotoxicity.
• Findings underscore the need for careful plasticizer selection and risk assessment to mitigate environmental and health hazards.