Use of a large dataset to develop new models for estimating the sorption of active pharmaceutical ingredients in soils and sediments

Jun Li; John L Wilkinson; Alistair B A Boxall

doi:10.1016/j.jhazmat.2021.125688

Use of a large dataset to develop new models for estimating the sorption of active pharmaceutical ingredients in soils and sediments

Jun Li, John L Wilkinson, Alistair B A Boxall

Environment and Geography

Research output: Contribution to journal › Article › peer-review

Abstract

Information on the sorption of active pharmaceutical ingredients (APIs) in soils and sediments is needed for assessing the environmental risks of these substances yet these data are unavailable for many APIs in use. Predictive models for estimating sorption could provide a solution. The performance of existing models is, however, often poor and most models do not account for the effects of soil/sediment properties which are known to significantly affect API sorption. Therefore, here, we use a high-quality dataset on the sorption behavior of 54 APIs in 13 soils and sediments to develop new models for estimating sorption coefficients for APIs in soils and sediments using three machine learning approaches (artificial neural network, random forest and support vector machine) and linear regression. A random forest-based model, with chemical and solid descriptors as the input, was the best performing model. Evaluation of this model using an independent sorption dataset from the literature showed that the model was able to predict sorption coefficients of 90% of the test set to within a factor of 10 of the experimental values. This new model could be invaluable in assessing the sorption behavior of molecules that have yet to be tested and in landscape-level risk assessments.

Original language	English
Article number	125688
Pages (from-to)	125688
Journal	Journal of hazardous materials
Volume	415
Early online date	24 Mar 2021
DOIs	https://doi.org/10.1016/j.jhazmat.2021.125688
Publication status	Published - 5 Aug 2021

Bibliographical note

Keywords

Adsorption
Geologic Sediments
Pharmaceutical Preparations
Soil
Soil Pollutants/analysis

Access to Document

10.1016/j.jhazmat.2021.125688

Journal pre-proof
© 2021 Elsevier B.V. This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy.
Accepted author manuscript, 1.45 MBLicence: CC BY-NC-ND

Cite this

@article{80c487413c364270bb524f033efb9bd1,

title = "Use of a large dataset to develop new models for estimating the sorption of active pharmaceutical ingredients in soils and sediments",

abstract = "Information on the sorption of active pharmaceutical ingredients (APIs) in soils and sediments is needed for assessing the environmental risks of these substances yet these data are unavailable for many APIs in use. Predictive models for estimating sorption could provide a solution. The performance of existing models is, however, often poor and most models do not account for the effects of soil/sediment properties which are known to significantly affect API sorption. Therefore, here, we use a high-quality dataset on the sorption behavior of 54 APIs in 13 soils and sediments to develop new models for estimating sorption coefficients for APIs in soils and sediments using three machine learning approaches (artificial neural network, random forest and support vector machine) and linear regression. A random forest-based model, with chemical and solid descriptors as the input, was the best performing model. Evaluation of this model using an independent sorption dataset from the literature showed that the model was able to predict sorption coefficients of 90% of the test set to within a factor of 10 of the experimental values. This new model could be invaluable in assessing the sorption behavior of molecules that have yet to be tested and in landscape-level risk assessments.",

keywords = "Adsorption, Geologic Sediments, Pharmaceutical Preparations, Soil, Soil Pollutants/analysis",

author = "Jun Li and Wilkinson, {John L} and Boxall, {Alistair B A}",

year = "2021",

month = aug,

day = "5",

doi = "10.1016/j.jhazmat.2021.125688",

language = "English",

volume = "415",

pages = "125688",

journal = "Journal of hazardous materials",

issn = "0304-3894",

publisher = "Elsevier",

}

TY - JOUR

T1 - Use of a large dataset to develop new models for estimating the sorption of active pharmaceutical ingredients in soils and sediments

AU - Li, Jun

AU - Wilkinson, John L

AU - Boxall, Alistair B A

PY - 2021/8/5

Y1 - 2021/8/5

N2 - Information on the sorption of active pharmaceutical ingredients (APIs) in soils and sediments is needed for assessing the environmental risks of these substances yet these data are unavailable for many APIs in use. Predictive models for estimating sorption could provide a solution. The performance of existing models is, however, often poor and most models do not account for the effects of soil/sediment properties which are known to significantly affect API sorption. Therefore, here, we use a high-quality dataset on the sorption behavior of 54 APIs in 13 soils and sediments to develop new models for estimating sorption coefficients for APIs in soils and sediments using three machine learning approaches (artificial neural network, random forest and support vector machine) and linear regression. A random forest-based model, with chemical and solid descriptors as the input, was the best performing model. Evaluation of this model using an independent sorption dataset from the literature showed that the model was able to predict sorption coefficients of 90% of the test set to within a factor of 10 of the experimental values. This new model could be invaluable in assessing the sorption behavior of molecules that have yet to be tested and in landscape-level risk assessments.

AB - Information on the sorption of active pharmaceutical ingredients (APIs) in soils and sediments is needed for assessing the environmental risks of these substances yet these data are unavailable for many APIs in use. Predictive models for estimating sorption could provide a solution. The performance of existing models is, however, often poor and most models do not account for the effects of soil/sediment properties which are known to significantly affect API sorption. Therefore, here, we use a high-quality dataset on the sorption behavior of 54 APIs in 13 soils and sediments to develop new models for estimating sorption coefficients for APIs in soils and sediments using three machine learning approaches (artificial neural network, random forest and support vector machine) and linear regression. A random forest-based model, with chemical and solid descriptors as the input, was the best performing model. Evaluation of this model using an independent sorption dataset from the literature showed that the model was able to predict sorption coefficients of 90% of the test set to within a factor of 10 of the experimental values. This new model could be invaluable in assessing the sorption behavior of molecules that have yet to be tested and in landscape-level risk assessments.

KW - Adsorption

KW - Geologic Sediments

KW - Pharmaceutical Preparations

KW - Soil

KW - Soil Pollutants/analysis

U2 - 10.1016/j.jhazmat.2021.125688

DO - 10.1016/j.jhazmat.2021.125688

M3 - Article

C2 - 34088186

SN - 0304-3894

VL - 415

SP - 125688

JO - Journal of hazardous materials

JF - Journal of hazardous materials

M1 - 125688

ER -