A new biodegradation prediction model specific to petroleum hydrocarbons

Area of Science:

• Environmental Chemistry
• Computational Chemistry
• Chemical Engineering

Background:

• Petroleum hydrocarbons are persistent environmental pollutants.
• Accurate prediction of biodegradation rates is crucial for environmental risk assessment.
• Existing models may not fully capture the complexity of hydrocarbon biodegradation.

Purpose of the Study:

• To develop a novel predictive model for quantitative primary biodegradation half-lives of individual petroleum hydrocarbons.
• To improve the accuracy of biodegradation half-life estimations for environmental applications.

Main Methods:

• A fragment-based approach using multiple linear regression was employed.
• The model utilizes counts of 31 distinct molecular fragments as descriptors.
• A dataset of 175 compounds with experimental biodegradation data was used for training and validation.

Main Results:

• The developed model achieved high accuracy, with R-squared values of 0.91 for the training set and 0.81 for the validation set.
• The model incorporates specific fragments to better represent ring systems in petroleum hydrocarbons and address data nonlinearity.
• The predictive performance demonstrates the model's utility for estimating biodegradation half-lives.

Conclusions:

• The new predictive model offers a reliable method for quantifying petroleum hydrocarbon biodegradation half-lives.
• This approach enhances the assessment of environmental fate and persistence of petroleum contaminants.
• The fragment-based strategy provides a valuable tool for environmental scientists and engineers.