Structure-biodegradability Relationships Research Articles

Meta-Analysis and Machine Learning Models for Anaerobic Biodegradation Rates of Organic Contaminants in Sediments and Sludge.

Anaerobic biodegradation rates (half-lives) of organic chemicals are pivotal for environmental risk assessment and remediation. Traditional experimental evaluation, constrained by prolonged, oxygen-free conditions, struggles to keep pace with emerging contaminants. Data-driven machine learning (ML) models serve as promising complements. However, reported quantitative structure-biodegradation relationships or ML models on anaerobic biodegradation are mostly based on small data sets (<100 records) and neglect experimental conditions, usually achieving compromised predictions. This work aimed to develop ML models for predicting the biodegradation half-lives of organic pollutants in anaerobic environments (i.e., sediment/soil and sludge). Focusing on important features of both chemicals and experimental conditions, we first curated two data sets, one for sediment/soil (SED) and the other for sludge (SLD), covering 978 records for 206 chemicals from the literature, and then conducted a meta-analysis. Next, we built a binary classification (half-life of 30 days as the cutoff) model with an accuracy of 81% and a regression model with R2 of 0.56 for SED based on LightGBM (80% and 0.31 for SLD based on Extra tree, respectively). The model interpretations underscored the significance of experimental conditions (e.g., temperature and inoculum dosage), as evidenced by their high feature importance, and the models were found to correctly capture the effects of chemical substructures, for example, branched structures and aromatic rings prolonged half-lives while methyl group and ortho-substitution on rings shortened half-lives. The applicability domains of the models were also defined, resulting in reasonable prediction for the half-lives of 41% (SED) or 67% (SLD) of over 4000 persistent, bioaccumulative, and toxic chemicals. Overall, this study pioneers ML models for predicting the anaerobic degradation half-lives, offering valuable support for future evaluation and implementation of chemical anaerobic biodegradation.

Improving predictions and understanding of primary and ultimate biodegradation rates with machine learning models

This study aimed to develop machine learning based quantitative structure biodegradability relationship (QSBR) models for predicting primary and ultimate biodegradation rates of organic chemicals, which are essential parameters for environmental risk assessment. For this purpose, experimental primary and ultimate biodegradation rates of high consistency were compiled for 173 organic compounds. A significant number of descriptors were calculated with a collection of quantum/computational chemistry software and tools to achieve comprehensive representation and interpretability. Following a pre-screening process, multiple QSBR models were developed for both primary and ultimate endpoints using three algorithms: extreme gradient boosting (XGBoost), support vector machine (SVM), and multiple linear regression (MLR). Furthermore, a unified QSBR model was constructed using the knowledge transfer technique and XGBoost. Results demonstrated that all QSBR models developed in this study had good performance. Particularly, SVM models exhibited high level of goodness of fit (coefficient of determination on the training set of 0.973 for primary and 0.980 for ultimate), robustness (leave-one-out cross-validated coefficient of 0.953 for primary and 0.967 for ultimate), and external predictive ability (external explained variance of 0.947 for primary and 0.958 for ultimate). The knowledge transfer technique enhanced model performance by learning from properties of two biodegradation endpoints. Williams plots were used to visualize the application domains of the models. Through SHapley Additive exPlanations (SHAP) analysis, this study identified key features affecting biodegradation rates. Notably, MDEO-12, APC2D1_C_O, and other features contributed to primary biodegradation, while AATS0v, AATS2v, and others inhibited it. For ultimate biodegradation, features like No. of Rotatable Bonds, APC2D1_C_O, and minHBa were contributors, while C1SP3, Halogen Ratio, GGI4, and others hindered the process. Also, the study quantified the contributions of each feature in predictions for individual chemicals. This research provides valuable tools for predicting both primary and ultimate biodegradation rates while offering insights into the mechanisms.

Aerobic Biotransformation and Defluorination of Fluoroalkylether Substances (ether PFAS): Substrate Specificity, Pathways, and Applications.

Fluoroalkylether substances (ether PFAS) constitute a large group of emerging PFAS with uncertain environmental fate. Among them, GenX is the well-known alternative to perfluorooctanoic acid and one of the six proposed PFAS to be regulated by the U.S. Environmental Protection Agency. This study investigated the structure-biodegradability relationship for 12 different ether PFAS with a carboxylic acid headgroup in activated sludge communities. Only polyfluorinated ethers with at least one -CH2- moiety adjacent to or a C=C bond in the proximity of the ether bond underwent active biotransformation via oxidative and hydrolytic O-dealkylation. The bioreactions at ether bonds led to the formation of unstable fluoroalcohol intermediates subject to spontaneous defluorination. We further demonstrated that this aerobic biotransformation/defluorination could complement the advanced reduction process in a treatment train system to achieve more cost-effective treatment for GenX and other recalcitrant perfluorinated ether PFAS. These findings provide essential insights into the environmental fate of ether PFAS, the design of biodegradable alternative PFAS, and the development of cost-effective ether PFAS treatment strategies.

High Persistence of Novel Polyfluoroalkyl Betaines in Aerobic Soils.

Some contemporary aqueous film-forming foams (AFFFs) contain n:3 and n:1:2 fluorotelomer betaines (FTBs), which are often detected at sites impacted by AFFFs. As new chemical replacements, little is known about their environmental fate. For the first time, we investigated the biotransformation potential of 5:3 and 5:1:2 FTBs and a commercial AFFF that mainly contains n:3 and n:1:2 FTBs (n = 5, 7, 9, 11, and 13). Although some polyfluoroalkyl compounds are precursors to perfluoroalkyl acids, 5:3 and 5:1:2 FTBs exhibited high persistence, with no significant changes even after 120 days of incubation. While the degradation of 5:3 FTB into suspected products such as fluorotelomer acids or perfluoroalkyl carboxylic acids (PFCAs) could not be conclusively confirmed, we did identify a potential biotransformation product, 5:3 fluorotelomer methylamine. Similarly, 5:1:2 FTB did not break down or produce short-chain hydrogen-substituted polyfluoroalkyl acids (n:2 H-FTCA), hydrogen-substituted PFCA (2H-PFCA), or any other products. Incubating the AFFF in four soils with differing properties and microbial communities resulted in 0.023-0.25 mol % PFCAs by day 120. Most of the products are believed to be derived from n:2 fluorotelomers, minor components of the AFFF. Therefore, the findings of the study cannot be fully explained by the current understanding of structure-biodegradability relationships.

Can AI Help Improve Water Quality? Towards the Prediction of Degradation of Micropollutants in Wastewater.

Micropollutants have become a serious environmental problem by threatening ecosystems and the quality of drinking water. This account investigates if advanced AI can be used to find solutions for this problem. We review background, the challenges involved, and the current state-of-the-art of quantitative structure-biodegradation relationships (QSBR). We report on recent progress combining experiment, quantum chemistry (QC) and chemoinformatics, and provide a perspective on potential future uses of AI technology to help improve water quality.

Identification of structure–biodegradability relationships for ionic liquids – clustering of a dataset based on structural similarity

In silico cluster analysis of the ionic liquids’ dataset revealed new insights into structure–biodegradability relationships for improved biodegradability to design fully mineralising ionic liquids (benign by design).

Primary and ultimate degradation of benzophenone-type UV filters under different environmental conditions and the underlying structure-biodegradability relationships

Ten common benzophenone-based UV filters (BPs), sharing the same basic structure and differing only in their substituents, were investigated with respect to their primary and ultimate biodegradability. This study was carried out in order to gain deeper insights into the relationship between structure and biodegradability. The primary biodegradation of the selected BPs was studied in river water at environmentally relevant concentrations (1 µg/L) while varying specific, crucial environmental conditions (aerobic, suboxic, supplementation of nutrients). For this purpose, both batch and column degradation tests were performed, which allowed a systematic study of the effects. Subsequently, the ultimate biodegradation, i.e. the potential to achieve full mineralization of BPs, was examined according to OECD guideline 301 F. The results indicate that mineralization is limited to derivatives in which both aromatic rings contain substituents. This hypothesis was supported by docking simulations showing systematic differences in the orientation of BPs within the active site of the cytochrome P450 enzyme. These differences in orientation correspond to the substitution pattern of the BPs. This study provides valuable insights for assessing the environmental hazards of this class of trace organic compounds.

An improved opposition-based crow search algorithm for biodegradable material classification

ABSTRACT The development of a reliable quantitative structure–activity relationship (QSAR) classification model with a small number of molecular descriptors is a crucial step in chemometrics. In this study, an improvement of crow search algorithm (CSA) is proposed by adapting the opposite-based learning (OBL) approach, which is named as OBL-CSA, to improve the exploration and exploitation capability of the CSA in quantitative structure–biodegradation relationship (QSBR) modelling of classifying the biodegradable materials. The results reveal that the performance of OBL-CSA not only manifest in improving the classification performance, but also in reduced computational time required to complete the process when compared to the standard CSA and other four optimization algorithms tested, which are the particle swarm algorithm (PSO), black hole algorithm (BHA), grey wolf algorithm (GWA), and whale optimization algorithm (WOA). In conclusion, the OBL-CSA could be a valuable resource in the classification of biodegradable materials.

Structure–Biodegradability Relationship of Nonylphenol Isomers in Two Soils with Long-Term Reclaimed Water Irrigation

Nonylphenol (NP), as one of the typical endocrine disrupter chemicals (EDCs), has a high detection concentration and frequency in reclaimed water. This research focused on the degradation of NP isomers in two typical reclaimed water irrigation fields in Daxing, China, and Florida, USA. The results showed that the half-lives of NP isomer degradation in the soil of China and Florida were 2.03–8.66 d and 5.16–11.83 d, respectively. The degradation of NP isomers was structure-specific. Isomers of NP5, NP2, NP11, and NP3 had the highest degradation rates in the two soils; NP12, NP7, and NP6 were the isomers with medium degradation rates; and NP4, NP1, NP10, NP9, and NP8 had the slowest degradation rates. Steric hindrance and mean information index for the magnitude of distance (IDWbar) were found to be the better indexes for measuring the degradation of NP isomers compared with the length of the side chain, the type of the substitute, and the molecular connectivity. This study offers insights into the characteristics of NP isomers and two reliable indicators for measuring the degradation of NP isomers, which could provide data support for the environmental fate and the health risk assessment of NP in the future.

Microbial Defluorination of Unsaturated Per- and Polyfluorinated Carboxylic Acids under Anaerobic and Aerobic Conditions: A Structure Specificity Study.

The recently discovered microbial reductive defluorination of two C6 branched and unsaturated fluorinated carboxylic acids (FCAs) provided valuable insights into the environmental fate of per- and polyfluoroalkyl substances (PFASs) and potential bioremediation strategies. However, a systematic investigation is needed to further demonstrate the role of C=C double bonds in the biodegradability of unsaturated PFASs. Here, we examined the structure-biodegradability relationships of 13 FCAs, including nine commercially available unsaturated FCAs and four structurally similar saturated ones, in an anaerobic defluorinating enrichment and an activated sludge community. The anaerobic and aerobic transformation/defluorination pathways were elucidated. The results showed that under anaerobic conditions, the α,β-unsaturation is crucial for FCA biotransformation via reductive defluorination and/or hydrogenation pathways. With sp2 C–F bonds being substituted by C–H bonds, the reductive defluorination became less favorable than hydrogenation. Moreover, for the first time, we reported enhanced degradability and defluorination capability of specific unsaturated FCA structures with trifluoromethyl (−CF3) branches at the α/β-carbon. Such FCA structures can undergo anaerobic abiotic defluorination in the presence of reducing agents and significant aerobic microbial defluorination. Given the diverse applications and emerging concerns of fluorochemicals, this work not only advances the fundamental understanding of the fate of unsaturated PFASs in natural and engineered environments but also may provide insights into the design of readily degradable fluorinated alternatives to existing PFAS compounds.

Towards the design of organosilicon compounds for environmental degradation by using structure biodegradability relationships

Organosilicon compounds have numerous applications in consumer products. After entering the environment most of them are resistant against microbial degradation and they persist in the environment. Accordingly, they are ubiquitously present in the environment.Therefore, better environmentally degradable organosilicon compounds are urgently needed. A systematic investigation of environmental degradability of organosilicon compounds allows to derive some general design principles, which in turn would enable chemists to reduce or better avoid environmental persistence of organosilicon compounds in the environment. Therefore, in this study, all organosilicon substances registered in the European Chemicals Agency (ECHA) database were evaluated for their environmental biodegradability. Results of own experiments with different organosilicon substances were added to extend the data basis. A dataset was generated. An assessment of all data was done and invalid data were excluded. The remaining 182 substances were grouped regarding their structure to derive general rules for the environmental biodegradability of organosilicon compounds. Non-biodegradable at all were for example cyclic, linear and branched siloxanes. Groups like ethers, esters, oximes, amines, and amides were prone to hydrolysis, which can result in readily biodegradable intermediates if they do not contain silicon functional groups anymore. This knowledge could be used for the design of better degradable organosilicon compounds as non-degradable substances should be avoided if they enter the environment after their usage.

Study on quantitative structure-biodegradability relationships of amine collectors by GFA-ANN method

In this study, the biodegradability of 17 amine collectors, categorized by fatty amine, quaternary ammonium compounds and oxygen-containing amine collectors, are tested with the Closed Bottle Test for 90 days, and the results indicate most amine collectors are not readily biodegradable. The oxygen-containing amine collectors have the best biodegradability due to the introduced oxygen-containing functional groups, subsequently fatty amine collectors with branched chains, while the tested quaternary ammonium compounds all have poor biodegradation ability. Besides, we search for and calculate 35 molecular descriptors to develop the quantitative structure biodegradability relationship (QSBR) of amine collectors. With the Genetic Function Approximation (GFA) algorithm, two sets of important molecular descriptors related to biodegradability (q) of amine collectors are selected from 35 molecular descriptors. Based on internal and external validations, the robust and reliable non-linear QSBR model with the squared correlation coefficient above 0.99 is determined via Artificial Neural Network (ANN) method, where the descriptors are respectively CL, N, ELUMO, δv2, indicating the biodegradable ability of amine collectors is correlated with the alkyl chain lengths (CL), the number of nitrogen atom-containing compounds (N), energy of the lowest unoccupied molecular orbital (ELUMO) and valence second-order connectivity index (δv2).

Using structure biodegradability relationships for environmentally benign design of organosilicons – An experimental comparison of organosilicons and their carbon analogues

Organosilicon substances are ubiquitous in the environment due to their stability and numerous applications in consumer products. Therefore, it is desirable to reduce their environmental persistency. Our study aimed to better understand the impact of silicon atoms in organic compounds on their environmental biodegradability as a contribution to sustainable chemistry. Accordingly, we investigated the biodegradability of organosilicon compounds and their carbon analogues. OECD 301D test was used to assess ready biodegradability. In addition, GC-MS analyses were performed to study the fate of the compounds in the test. Three out of five carbon compounds and no organosilicon compound were found readily biodegradable. In all but one case, higher biodegradation degrees could be observed for the carbon compounds. Hydrolysis was identified as a mandatory step prior to the biodegradation of organosilicon substances. The silicon-free product of hydrolysis determined the rate of biodegradation. The silicon-containing reaction products of hydrolysis were not biodegradable. The high biodegradability of one organosilicon compound can be attributed to faster hydrolysis due to an easily hydrolysable Si–N bond and a high biodegradation rate of the resulting silicon-free hydrolysis product. Insertion of such heteroatoms or functional groups into polysiloxane chains may be a promising approach to benign organosilicon compounds.

Comparison of Small Molecule Biotransformation Half-Lives between Activated Sludge and Soil: Opportunities for Read-Across?

Compartment-specific degradation half-lives are essential pieces of information in the regulatory risk assessment of synthetic chemicals. However, their measurement according to regulatory testing guidelines is laborious and costly. Despite the obvious ecological and economic benefits of knowing environmental degradability as early as possible, its consideration in the early phases of rational chemical design is therefore challenging. Here, we explore the possibility to use half-lives determined in highly time- and work-efficient biotransformation experiments with activated sludge and mixtures of chemicals to predict soil half-lives from regulatory simulation studies. We experimentally determined half-lives for 52 structurally diverse agrochemical active ingredients in batch reactors with three concentrations of the same activated sludge. We then developed bi- and multivariate models for predicting half-lives in soil by regressing the experimentally determined half-lives in activated sludge against average soil half-lives of the same chemicals extracted from regulatory data. The models differed in how we accounted for sorption-related bioavailability differences in soil and activated sludge. The best-performing models exhibited good coefficients of determination (R2 of around 0.8) and low average errors (<factor of 3 in half-life predictions) and were robust in cross-validation. From a practical perspective, these results suggest that it may indeed be possible to read across from half-lives determined in highly efficient biotransformation experiments in activated sludge to soil half-lives, which are obtained from much more work- and resource-intense regulatory studies, and that these predictions are clearly superior to predictions based on the output of BIOWIN, a publicly available quantitative structure-biodegradation relationship (QSBR) model. From a theoretical perspective, these results suggest that soil and activated sludge microbial communities, although certainly different in terms of taxonomic composition, may be functionally similar with respect to the enzymatic transformation of environmentally relevant concentrations of a diverse range of chemical compounds.

Disentanglement of the chemical, physical, and biological processes aids the development of quantitative structure-biodegradation relationships for aerobic wastewater treatment

Attenuation of organic compounds in sewage treatment plants (STPs) is affected by a complex interplay between chemical (e.g. ionization, hydrolysis), physical (e.g. sorption, volatilization), and biological (e.g. biodegradation, microbial acclimation) processes. These effects should be accounted for individually, in order to develop predictive cheminformatics tools for STPs. Using measured data from 70 STPs in the Netherlands for 69 chemicals (pharmaceuticals, herbicides, etc.), we highlighted the influences of 1) chemical ionization, 2) sorption to sludge, and 3) acclimation of the microbial consortia on the primary removal of chemicals. We used semi-empirical corrections for each of these influences to deduce biodegradation rate constants upon which quantitative structure-biodegradation relationships (QSBRs) were developed. As shown by a global QSBR, biodegradation in STPs generally relates to structural complexity, size, energetics, and charge distribution. Statistics of the global QSBR were reasonable, being R2training=0.69 (training set of 51 compounds) and R2validation=0.50 (validation set of 18 compounds). Class-specific QSBRs utilized electronic properties potentially relating to rate-limiting enzymatic steps. For class-specific QSBRs, values of R2 of in between 0.7 and 0.8 were obtained. With caution, environmental risk assessment methodologies may apply these models to estimate biodegradation rates for ‘data-poor’ compounds. The approach also highlights ‘meta data’ on STP operational parameters needed to develop QSBRs of better predictability in the future.

The experimental determination of reliable biodegradation rates for mono-aromatics towards evaluating QSBR models

Quantitative Structure Biodegradation Relationships (QSBRs) are a tool to predict the biodegradability of chemicals. The objective of this work was to generate reliable biodegradation data for mono-aromatic chemicals in order to evaluate and verify previously developed QSBRs models. A robust biodegradation test method was developed to estimate specific substrate utilization rates, which were used as a proxy for biodegradation rates of chemicals in pure culture. Five representative mono-aromatic chemicals were selected that spanned a wide range of biodegradability. Aerobic biodegradation experiments were performed for each chemical in batch reactors seeded with known degraders. Chemical removal, degrader growth and CO2 production were monitored over time. Experimental data were interpreted using a full carbon mass balance model, and Monod kinetic parameters (Y, Ks, qmax and μmax) for each chemical were determined. In addition, stoichiometric equations for aerobic mineralization of the test chemicals were developed. The theoretically estimated biomass and CO2 yields were similar to those experimentally observed; 35% (s.d ± 8%) of the recovered substrate carbon was converted to biomass, and 65% (s.d ± 8%) was mineralised to CO2. Significant correlations were observed between the experimentally determined specific substrate utilization rates, as represented by qmax and qmax/Ks, at high and low substrate concentrations, respectively, and the first order biodegradation rate constants predicted by a previous QSBR study. Similarly, the correlation between qmax and selected molecular descriptors characterizing the chemicals structure in a previous QSBR study was also significant. These results suggest that QSBR models can be reliable and robust in prioritising chemical half-lives for regulatory screening purposes.

A quantitative structure-biodegradation relationship (QSBR) approach to predict biodegradation rates of aromatic chemicals

The objective of this work was to develop a QSBR model for the prioritization of organic pollutants based on biodegradation rates from a database containing globally harmonized biodegradation tests using relevant molecular descriptors. To do this, we first categorized the chemicals into three groups (Group 1: simple aromatic chemicals with a single ring, Group 2: aromatic chemicals with multiple rings and Group3: Group 1 plus Group 2) based on molecular descriptors, estimated the first order biodegradation rate of the chemicals using rating values derived from the BIOWIN3 model, and finally developed, validated and defined the applicability domain of models for each group using a multiple linear regression approach. All the developed QSBR models complied with OECD principles for QSAR validation. The biodegradation rate in the models for the two groups (Group 2 and 3 chemicals) are associated with abstract molecular descriptors that provide little relevant practical information towards understanding the relationship between chemical structure and biodegradation rates. However, molecular descriptors associated with the QSBR model for Group 1 chemicals (R2 = 0.89, Q2loo = 0.87) provided information on properties that can readily be scrutinised and interpreted in relation to biodegradation processes. In combination, these results lead to the conclusion that QSBRs can be an alternative tool to estimate the persistence of chemicals, some of which can provide further insights into those factors affecting biodegradation.

Prediction of Biodegradability for Polycyclic Aromatic Hydrocarbons Using Various In Silico Modeling Methods.

Polycyclic aromatic hydrocarbons (PAHs) have attracted great concern as global environmental pollutants. In this work, the quantitative structure-biodegradability relationship (QSBR) study has been done to predict the biodegradability of PAHs and develop the correlation between the biodegradability and the molecular structures. The structural chemistry and quantum chemistry descriptors were used to represent molecular structures. Three in silico modeling methods, i.e., multiple linear regression (MLR), radial basis function neural network, and back-propagation artificial neural network (BPANN), are utilized to construct the linear and nonlinear prediction models and provide some insights into the structural characteristics affecting the biodegradability of PAHs. The stability of these QSBR models was tested by leave-one-out cross-validation, and the cross-validated correlation coefficients (q2) were 0.6109, 0.6887, and 0.6586, respectively. The correlation coefficients (R2) of the three models for the training set were 0.7811, 0.8883, and 0.9667, respectively. The comparison of the three models showed that the BPANN model produced a statistically more significant model than the other two models. On the basis of molecular structure, the dominant molecular structure descriptor affecting biodegradability of PAHs were analyzed and discussed.

Ultrasonic-assisted biodegradation of endocrine disrupting compounds by Pseudomonas putida the importance of rhamnolipid for intermediate product degradation

The present study aimed to completely remove estrogens, including oestrone(E1), oestradiol(E2), oes-triol(E3), 17α-ethinylestradiol(EE2) and bisphenol-A(BPA), from soil using Pseudomonas putida(P. putida). A central composite design was developed to determine the optimal conditions of three variables(ultrasonication time, quantity of P. putida, and concentration of added rhamnolipid) for the removal of the estrogens, and the biodegradation rates of the estrogens were investigated under the optimum conditions. Moreover, a quantitative structure-biodegradation relationship(QSBR) was used to analyze the effect of the estrogenic physicochemical properties on the enhancement of the biological degradation. The optimal conditions were an ultrasonication time of 3 min, a P. putida quantity of 8 mL, and a rhamnolipid concentration of 100 mg/L. These conditions resulted in removal of 100%, 94.86%, 94.90%, 96.56% and 94.56% of E1, E2, EE2, BPA and E3, respectively after 7 d. The degradations were more rapid and com-plete than those reported in previous studies, indicating the suitability of the adaptation of P. putida to estrogen de-gradation under conditions of ultrasonic-assistance and adding rhamnolipid; improvement was particularly apparent from the complete degradation of E3. Based on a Pearson correlation analysis, the estrogen molecule polar surface area(PSA) and surface tension were significantly related to the biodegradation effect. An analysis of the QSBR model with the estrogen biodegradation rates as a dependent variable and the PSA and surface tension as independent va-riables indicated that larger PSA caused decreased estrogen biodegradation, while the biodegradation progress was dominated by the surface tension of the estrogens. The interaction of PSA and surface tension had an antagonistic ef-fect on the biodegradation of estrogens. Therefore, rhamnolipid/ultrasonication can significantly improve the biode-gradation rates of oestrogens in soil, while simultaneously adjusting other environmental conditions would influence and control the biodegradation processes of estrogens.

Development of QSBR models for anoxic biodegradability of polycyclic aromatic hydrocarbons by using SMLR and BP-ANN

Abstract In the present study, quantitative structure-biodegradability relationship (QSBR) models were developed for the anoxic Andrews model parameters of polycyclic aromatic hydrocarbons (PAHs). The molecular geometries of 20 PAHs were studied using density functional theory. Stepwise multiple linear regression (SMLR) and backpropagation artificial neural network (BP-ANN) methods were applied to establish the QSBR model between the anoxic biodegradability (q max ) and 5 molecular descriptors. Results of the regression analysis indicated that both the accuracy and predictive ability of the BP-ANN model were better than those of SMLR model. After analyzing the sensitivity of variables, the key molecular structure descriptor influencing anoxic biodegradability of PAHs were screened to be E HOMO and Freq. The present study demonstrates the value of QSBR not only as a predictive tool but also as a framework for understanding the mechanisms governing biodegradation at the molecular level.