Data leakage management enables reliable machine-learning-based prediction of organic pollutant adsorption onto microplastics

Abstract

Machine learning (ML) has been increasingly used to model the adsorption of organic pollutants (OPs) onto microplastics (MPs). However, existing studies have not adequately captured key influencing factors or implemented proper data leakage management (DLM) during model development. To bridge these gaps, ML models were developed using comprehensive features including microplastic properties, pollutant properties, and adsorption conditions, and three DLM strategies, namely data point-, isotherm-, and study-based, were evaluated on datasets. Furthermore, feature-similarity analysis provided mechanistic support for isotherm-based data partitioning, extending model selection beyond predictive performance alone. The resulting models achieved robust predictive performance and revealed the mechanistic relevance of comprehensive factors integration, with equilibrium concentration and hydrogen bonding emerging as dominant drivers. Overall, by addressing the critical but previously overlooked issue of data leakage and underscoring the need for holistic feature integration, this study establishes a new methodological baseline that resolves key limitations in current MP–OP adsorption modeling and guides future research.

Graphical abstract

Keywords

Microplastics / Organic pollutants / Adsorption / Machine learning / Data leakage management

Highlight

	● Isotherm DLM prevents overoptimistic predictions for OP adsorption on MPs.
	● XGB delivers reliable performance with realistic external generalization.
	● Cosine similarity quantifies train–test overlap and confirms DLM efficacy.
	● Model interpretation identifies log C e and hydrogen bonding as key drivers.
	● This framework enables reproducible, leakage-aware modeling of MP–OP systems.

References

[1] Abiodun O I , Jantan A , Omolara A E , Dada K V , Umar A M , Linus O U . (2019). Comprehensive review of artificial neural network applications to pattern recognition. IEEE Access, 7: 158820–158846

[2] Anagu I , Ingwersen J , Utermann J , Streck T . (2009). Estimation of heavy metal sorption in German soils using artificial neural networks. Geoderma, 15(1−2): 104–112

[3] Apul O G , Wang Q L , Shao T , Rieck J R , Karanfil T . (2013). Predictive model development for adsorption of aromatic contaminants by multi-walled carbon nanotubes. Environmental Science & Technology, 47(5): 2295–2303

[4] Bischl B , Binder M , Lang M , Pielok T , Richter J , Coors S , Thomas J , Ullmann T , Becker M , Boulesteix A L . et al. (2023). Hyperparameter optimization: foundations, algorithms, best practices, and open challenges. WIREs Data Mining and Knowledge Discovery, 13(2): e1484

[5] Bouwmeester H , Hollman P C H , Peters R J B . (2015). Potential health impact of environmentally released micro-and nanoplastics in the human food production chain: experiences from nanotoxicology. Environmental Science & Technology, 49(15): 8932–8947

[6] Bryant M T , Ma X M . (2024). Machine learning prediction of adsorption behavior of xenobiotics on microplastics under different environmental conditions. ACS ES&T Water, 4(3): 991–999

[7] Cao L , Wu D , Liu P , Hu W Y , Xu L , Sun Y C , Wu Q M , Tian K , Huang B , Yoon S J . et al. (2021). Occurrence, distribution and affecting factors of microplastics in agricultural soils along the lower reaches of Yangtze River, China. Science of the Total Environment, 794: 148694

[8] Cheadle C , Vawter M P , Freed W J , Becker K G . (2003). Analysis of microarray data using Z score transformation. The Journal of Molecular Diagnostics, 5(2): 73–81

[9] Chen K Y , Chen H X , Zhou C L , Huang Y C , Qi X Y , Shen R Q , Liu F R , Zuo M , Zou X Y , Wang J F . et al. (2020). Comparative analysis of surface water quality prediction performance and identification of key water parameters using different machine learning models based on big data. Water Research, 171: 115454

[10] Cui W X , Hale R C , Huang Y C , Zhou F L , Wu Y , Liang X L , Liu Y , Tan H L , Chen D . (2023). Sorption of representative organic contaminants on microplastics: effects of chemical physicochemical properties, particle size, and biofilm presence. Ecotoxicology and Environmental Safety, 251: 114533

[11] Enyoh C E , Wang Q Y , Lu S L . (2023). Optimizing the efficient removal of ciprofloxacin from aqueous solutions by polyethylene terephthalate microplastics using multivariate statistical approach. Chemical Engineering Science, 278: 118917

[12] Fumagalli F , Muschalik M , Hüllermeier E , Hammer B . (2023). Incremental permutation feature importance (iPFI): towards online explanations on data streams. Machine Learning, 112(12): 4863–4903

[13] Gui B X , Xu X T , Zhang S N , Wang Y , Li C , Zhang D M , Su L M , Zhao Y H . (2021). Prediction of organic compounds adsorbed by polyethylene and chlorinated polyethylene microplastics in freshwater using QSAR. Environmental Research, 197: 111001

[14] Guo Y P , Shao W F , Zhao W G , Zhu H . (2024). The occurrence, distribution, environmental effects, and interactions of microplastics and antibiotics in the aquatic environment of China. Water, 16(10): 1435

[15] Ho W K , Law J C F , Zhang T , Leung K S Y . (2020). Effects of weathering on the sorption behavior and toxicity of polystyrene microplastics in multi-solute systems. Water Research, 187: 116419

[16] Kim C , Park T . (2022). Predicting determinants of lifelong learning intention using gradient boosting machine (GBM) with grid search. Sustainability, 14(9): 5256

[17] Lai M , Wulff S S , Cao Y T , Robinson T J , Rajapaksha R . (2023). An interpretable time series machine learning method for varying forecast and nowcast lengths in wastewater-based epidemiology. MethodsX, 11: 102382

[18] Li J , Zhang K N , Zhang H . (2018). Adsorption of antibiotics on microplastics. Environmental Pollution, 237: 460–467

[19] Li Z R , Li S S , Li L Z , Tao L , Zhu X H , Ma R X , Sun B B , Zhang Y H , Yu Y J . (2022). Interaction of tetrabromobisphenol A (TBBPA) with microplastics-sediment (MPs-S) complexes: a comparison between binary and simple systems. Environmental Pollution, 301: 118991

[20] Lu Z J , Macfarlane J K , Gschwend P M . (2016). Adsorption of organic compounds to diesel soot: frontal analysis and polyparameter linear free-energy relationship. Environmental Science & Technology, 50(1): 285–293

[21] Lundberg S M, Lee S I (2017). A unified approach to interpreting model predictions. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates Inc

[22] Ma J , Zhao J H , Zhu Z L , Li L Q , Yu F . (2019). Effect of microplastic size on the adsorption behavior and mechanism of triclosan on polyvinyl chloride. Environmental Pollution, 254: 113104

[23] Mato Y , Isobe T , Takada H , Kanehiro H , Ohtake C , Kaminuma T . (2001). Plastic resin pellets as a transport medium for toxic chemicals in the marine environment. Environmental Science & Technology, 35(2): 318–324

[24] Paleyes A , Urma R G , Lawrence N D . (2023). Challenges in deploying machine learning: a survey of case studies. ACM Computing Surveys, 55(6): 114

[25] Prajapati A , Narayan Vaidya A , Kumar A R . (2022). Microplastic properties and their interaction with hydrophobic organic contaminants: a review. Environmental Science and Pollution Research, 29(33): 49490–49512

[26] Qiu Y , Li Z J , Zhang T , Zhang P . (2023). Predicting aqueous sorption of organic pollutants on microplastics with machine learning. Water Research, 244: 120503

[27] Rahman M A , Hasan M R , Riya K K , Bhuyain M A B , Semme S A , Ngah N , Arai T , Hossain M B . (2026). Characteristics and risks of microplastic contamination in marine trash fish from the northern Bay of Bengal: implications for human health and local aquaculture industry. Engineering Environment, 20(8): 121

[28] Sengupta A , Keskar M , Jayabun S . (2016). Sorption behaviour of metal ion on thorium tungstate synthesized by solid state route. Journal of Radioanalytical and Nuclear Chemistry, 310(3): 979–989

[29] Shih C Y , Wang Y H , Chen Y J , Chen H A , Lin A Y C . (2021). Enhanced sorption of the UV filter 4-methylbenzylidene camphor on aged PET microplastics from both experimental and theoretical perspectives. RSC Advances, 11(51): 32494–32504

[30] Tang Z P , Zhu F Y , Jiang T Y , Wei F X , Gao Y Y , Xiang C , Duan Y , Hua Y L , Zhou S , Wang Y Y . (2023). Oxygen-containing functional groups enhance uranium adsorption by aged polystyrene microplastics: experimental and theoretical perspectives. Chemical Engineering Journal, 465: 142730

[31] Thompson R C, Olsen Y, Mitchell R P, Davis A, Rowland S J, John A W G, Mcgonigle D, Russell A E (2004). Lost at sea: where is all the plastic? Science, 304(5672): 838

[32] Uber T H , Hüffer T , Planitz S , Schmidt T C . (2019). Characterization of sorption properties of high-density polyethylene using the poly-parameter linearfree-energy relationships. Environmental Pollution, 248: 312–319

[33] Vidal M , Santos M J , Abrão T , Rodríguez J , Rigol A . (2009). Modeling competitive metal sorption in a mineral soil. Geoderma, 149(3−4): 189–198

[34] Wang F , Zhang M , Sha W , Wang Y D , Hao H Z , Dou Y Y , Li Y . (2020a). Sorption behavior and mechanisms of organic contaminants to nano and microplastics. Molecules, 25(8): 1827

[35] Wang J , Li J , Wang Q , Sun Y Z . (2020b). Microplastics as a vector for HOC bioaccumulation in earthworm Eisenia fetida in soil: importance of chemical diffusion and particle size. Environmental Science & Technology, 54(19): 12154–12163

[36] Wang J , Liu X H , Liu G N , Zhang Z X , Wu H , Cui B S , Bai J H , Zhang W . (2019). Size effect of polystyrene microplastics on sorption of phenanthrene and nitrobenzene. Ecotoxicology and Environmental Safety, 173: 331–338

[37] Wang J , Tao J G , Wu M C , Sun Y Z , Su Y , Guo X T , Du X Y , Li J , Gan J . (2024). Size-dependent vector effects of microplastics on bioaccumulation of hydrophobic organic contaminants in earthworm: a dual-dosing study. Environment International, 186: 108625

[38] Wang Y , Wang X J , Li Y , Li J , Liu Y Y , Xia S Q , Zhao J F . (2021). Effects of exposure of polyethylene microplastics to air, water and soil on their adsorption behaviors for copper and tetracycline. Chemical Engineering Journal, 404: 126412

[39] Xu J P , Zhang K , Wang L , Yao Y M , Sun H W . (2022). Strong but reversible sorption on polar microplastics enhanced earthworm bioaccumulation of associated organic compounds. Journal of Hazardous Materials, 423: 127079

[40] Yang C Y, Brower-Sinning R A, Lewis G, Kästner C (2023a). Data leakage in notebooks: static detection and better processes. In: Proceedings of the 37th IEEE/ACM International Conference on Automated Software Engineering. Rochester: Association for Computing Machinery

[41] Yang H R , Huang K , Zhang K , Weng Q , Zhang H C , Wang F E . (2021). Predicting heavy metal adsorption on soil with machine learning and mapping global distribution of soil adsorption capacities. Environmental Science & Technology, 55(20): 14316–14328

[42] Yang M Q , Zhu J J , McGaughey A , Zheng S X , Priestley R D , Ren Z J . (2023b). Predicting extraction selectivity of acetic acid in pervaporation by machine learning models with data leakage management. Environmental Science & Technology, 57(14): 5934–5946

[43] Yao J J , Li H P , Yang H Y . (2023). Predicting adsorption capacity of pharmaceuticals and personal care products on long-term aged microplastics using machine learning. Journal of Hazardous Materials, 458: 131963

[44] Yao J J , Wen J Y , Li H P , Yang Y . (2022). Surface functional groups determine adsorption of pharmaceuticals and personal care products on polypropylene microplastics. Journal of Hazardous Materials, 423: 127131

[45] Zhang K , Zhong S F , Zhang H C . (2020). Predicting aqueous adsorption of organic compounds onto biochars, carbon nano-tubes, granular activated carbons, and resins with machine learning. Environmental Science & Technology, 54(11): 7008–7018

[46] Zhao L F , Rong L L , Xu J P , Lian J P , Wang L , Sun H W . (2020). Sorption of five organic compounds by polar and nonpolar microplastics. Chemosphere, 257: 127206

[47] Zhong S F , Zhang K , Bagheri M , Burken J G , Gu A , Li B K , Ma X M , Marrone B L , Ren Z J , Schrier J . et al. (2021). Machine learning: new ideas and tools in environmental science and engineering. Environmental Science & Technology, 55(19): 12741–12754

[48] Zhu H , He J L , Wu Y M , Tong L Z , Zhang W H , Zhuang L W . (2024). Assessment of global antibiotic exposure risk for crops: incorporating soil adsorption via machine learning. Environ-mental Science & Technology, 58(30): 13327–13336

[49] Zhu J J , Yang M Q , Ren Z J . (2023). Machine learning in environmental research: common pitfalls and best practices. Environmental Science & Technology, 57(46): 17671–17689

[50] Zhu T Y , Chen W X , Cheng H M , Wang Y J , Singh R P . (2019). Prediction of polydimethylsiloxane-water partition coefficients based on the pp-LFER and QSAR models. Ecotoxicology and Environmental Safety, 182: 109374

Rights & permissions

Higher Education Press 2027