Low salinity influences the dose-dependent transcriptomic responses of oysters to cadmium.

Species in estuaries tend to undergo both cadmium (Cd) and low salinity stress. However, how low salinity affects the Cd toxicity has not been fully understood. Investigating the impacts of low salinity on the dose-response relationships between Cd and biological endpoints has potential to enhance our understanding of the combined effects of low salinity and Cd. In this work, changes in the transcriptomes of Pacific oysters were analyzed following exposure to Cd (5, 20, 80 µg/L Cd2+) under normal (31.4 psu) and low (15.7 psu) salinity conditions, and then the dose-response relationship between Cd and transcriptome was characterized in a high-throughput manner. The benchmark dose (BMD) of gene expression, as a point of departure (POD), was also calculated based on the fitted dose-response model. We found that low salinity treatment significantly influenced the dose-response relationships between Cd and transcripts in oysters indicated by altered dose-response curves. In details, a total of 219 DEGs were commonly fitted to best models under both normal and low salinity conditions. Nearly three quarters of dose-response curves varied with salinity condition. Some monotonic dose-response curves in normal salinity condition even were replaced by nonmonotonic curves in low salinity condition. Low salinity treatment decreased the PODs of differentially expressed genes induced by Cd, suggesting that gene differential expression was more prone to being triggered by Cd in low salinity condition. The changed sensitivity to Cd in low salinity condition should be taken into consideration when using oyster as an indicator to assess the ecological risk of Cd pollution in estuaries.

Time Is Ripe for Targeting Per- and Polyfluoroalkyl Substances-Induced Hormesis: Global Aquatic Hotspots and Implications for Ecological Risk Assessment.

Globally implemented ecological risk assessment (ERA) guidelines marginalize hormesis, a biphasic dose-response relationship characterized by low-dose stimulation and high-dose inhibition. The present study illuminated the promise of hormesis as a scientific dose-response model for ERA of per- and polyfluoroalkyl substances (PFAS) represented by perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS). A total of 266 hormetic dose-response relationships were recompiled from 1237 observations, covering 30 species from nine representative taxonomic groups. The standardized hormetic amplitudes followed the log-normal probability distribution, being subject to the limits of biological plasticity but independent of stress inducers. The SHapley Additive exPlanations algorithm revealed that the target endpoint was the most important variable explaining the hormetic amplitudes. Subsequently, quantitative frameworks were established to incorporate hormesis into the predicted no-effect concentration levels, with a lower induction dose and a zero-equivalent point but a broader hormetic zone for PFOS. Realistically, 10,117 observed concentrations of PFOA and PFOS were gathered worldwide, 4% of which fell within hormetic zones, highlighting the environmental relevance of hormesis. Additionally, the hormesis induction potential was identified in other legacy and emerging PFAS as well as their alternatives and mixtures. Collectively, it is time to incorporate the hormesis concept into PFAS studies to facilitate more realistic risk characterizations.

Global prevalence of microplastics in tap water systems: Abundance, characteristics, drivers and knowledge gaps.

Tap water is a main route for human direct exposure to microplastics (MPs). This study recompiled baseline data from 34 countries to assess the current status and drivers of MP contamination in global tap water systems (TWS). It was shown that MPs were detected in 87 % of 1148 samples, suggesting the widespread occurrence of MPs in TWS. The detected concentrations of MPs spanned seven orders of magnitude and followed the linearized log-normal distribution (MSE = 0.035, R2 = 0.965), with cumulative concentrations at 5th, 50th and 95th percentiles of 0.028, 4.491 and 728.105 items/L, respectively. The morphological characteristics were further investigated, indicating that particles smaller than 50 µm dominated in global TWS, with fragment, polyester and transparent as the most common shape, composition and color of MPs, respectively. Subsequently, the SHapley Additive exPlanations (SHAP) algorithm was implemented to quantify the importance of variables affecting the MP abundance in global TWS, showing that the lower particle size limit was the most important variables. Subgroup analysis revealed that the concentration of MPs counted at the size limit of 1 µm was >20 times higher than that above 1 µm. Ultimately, current knowledge gaps and future research needs were elucidated.

Regulating Cholesterol in Tumorigenesis: A Novel Paradigm for Tumor Nanotherapeutics.

During the past decade, "membrane lipid therapy", which involves the regulation of the structure and function of tumor cell plasma membranes, has emerged as a new strategy for cancer treatment. Cholesterol is an important component of the tumor plasma membrane and serves an essential role in tumor initiation and progression. This review elucidates the role of cholesterol in tumorigenesis (including tumor cell proliferation, invasion/metastasis, drug resistance, and immunosuppressive microenvironment) and elaborates on the potential therapeutic targets for tumor treatment by regulating cholesterol. More meaningfully, this review provides an overview of cholesterol-integrated membrane lipid nanotherapeutics for cancer therapy through cholesterol regulation. These strategies include cholesterol biosynthesis interference, cholesterol uptake disruption, cholesterol metabolism regulation, cholesterol depletion, and cholesterol-based combination treatments. In summary, this review demonstrates the tumor nanotherapeutics based on cholesterol regulation, which will provide a reference for the further development of "membrane lipid therapy" for tumors.

Risk assessment of trace metals and polycyclic aromatic hydrocarbons in seawater of typical bays in the Bohai Sea.

The ecological risks of trace metals (Cu, Zn, As, Cd, Pb, and Hg) and PAHs in seawater from three typical bays of the Bohai Sea (the Liaodong Bay, Bohai Bay, and Laizhou Bay) were comprehensively assessed by recompiling 637 sites. Results highlighted that scrutiny should be given to the ecological risks of Cu (3.80 µg/L) in the Bohai Bay and Hg (0.23 µg/L) in the Laizhou Bay. Conversely, the Liaodong Bay exhibited negligible ecological risks related to trace metals. The risks of ΣPAHs in the Liaodong Bay, Bohai Bay, and Laizhou Bay were moderate, with mean concentrations of 368.16 ng/L, 731.93 ng/L, and 187.58 ng/L, respectively. The source allocation of trace metals and PAHs required consideration of spatial variability and anthropogenic factors, which greatly affected the distribution and composition of these pollutants. The combined ecological risks in the Bohai Bay (6.80 %) and Laizhou Bay (5.43 %) deserved more attention.

RGDAN: A random graph diffusion attention network for traffic prediction.

Traffic Prediction based on graph structures is a challenging task given that road networks are typically complex structures and the data to be analyzed contains variable temporal features. Further, the quality of the spatial feature extraction is highly dependent on the weight settings of the graph structures. In the transportation field, the weights of these graph structures are currently calculated based on factors like the distance between roads. However, these methods do not take into account the characteristics of the road itself or the correlations between different traffic flows. Existing approaches usually pay more attention to local spatial dependencies extraction while global spatial dependencies are ignored. Another major problem is how to extract sufficient information at limited depth of graph structures. To address these challenges, we propose a Random Graph Diffusion Attention Network (RGDAN) for traffic prediction. RGDAN comprises a graph diffusion attention module and a temporal attention module. The graph diffusion attention module can adjust its weights by learning from data like a CNN to capture more realistic spatial dependencies. The temporal attention module captures the temporal correlations. Experiments on three large-scale public datasets demonstrate that RGDAN produces predictions with 2%-5% more precision than state-of-the-art methods.

The national, regional, and global impact of glaucoma as reported in the 2019 Global Burden of Disease Study.

Introduction: We investigated the disability-adjusted life years (DALYs) of glaucoma. Methods: The estimated annual percentage change (EAPC) was measured to assess trends in the age-standardized DALY rate from 1990 to 2019. Results: The global age-standardized DALY rate of glaucoma decreased with an EAPC of -1.00. The age-standardized DALY rate decreased least in high-SDI regions. Eastern sub-Saharan Africa had highest age-standardized DALY rate in 2019. At the national level, Mali had the highest age-standardized DALY rate in 2019. Conclusions: Although the global burden of glaucoma has decreased, the burden remain high in regions with low SDI values and in sub-Saharan Africa.

CENPE and LDHA were potential prognostic biomarkers of chromophobe renal cell carcinoma.

BACKGROUND: Most sarcomatoid differentiated renal cell carcinoma was differentiated from Chromophobe renal cell carcinoma (KICH) and related to a bad prognosis. Thus, finding biomarkers is important for the therapy of KICH. METHODS: The UCSC was used for determining the expression of mRNA and miRNA and clinical data in KICH and normal samples. KEGG and GO were used for predicting potential function of differently expressed genes (DEGs). Optimal prognostic markers were determined by Lasso regression. Kaplan-Meier survival, ROC, and cox regression were used for assessing prognosis value. GSEA was used for predicting potential function of markers. The relations between markers and immune cell infiltration were determined by Pearson method. The upstream miRNA of markers was predicted in TargetScan and DIANA. RESULTS: The 6162 upregulated and 13,903 downregulated DEGs were identified in KICH. Further CENPE and LDHA were screened out as optimal prognostic risk signatures. CENPE was highly expressed while LDHA was lowly expressed in KICH samples, and the high expressions of 2 genes contributed to bad prognosis. The functions of CENPE and LDHA were mainly enriched in proliferation related pathways such as cell cycle and DNA replication. In addition, the correlation of 2 genes with immune infiltrates in KICH was also observed. Finally, we found that has-miR-577 was the common upstream of 2 genes and the binding sites can be predicted. CONCLUSION: CENPE and LDHA were identified as the important prognostic biomarkers in KICH, and they might be involved in the proliferation of cancer cell.

An Adversarial Time-Frequency Reconstruction Network for Unsupervised Anomaly Detection.

Detecting anomalies in massive volumes of multivariate time series data, particularly in the IoT domain, is critical for maintaining stable systems. Existing anomaly detection models based on reconstruction techniques face challenges in distinguishing normal and abnormal samples from unlabeled data, leading to performance degradation. Moreover, accurately reconstructing abnormal values and pinpointing anomalies remains a limitation. To address these issues, we introduce the Adversarial Time-Frequency Reconstruction Network for Unsupervised Anomaly Detection (ATF-UAD). ATF-UAD consists of a time reconstructor, a frequency reconstructor and a dual-view adversarial learning mechanism. The time reconstructor utilizes a parity sampling mechanism to weaken the dependency between neighboring points. Then attention mechanisms and graph convolutional networks (GCNs) are used to update the feature information for each point, which combines points with close feature relationships and dilutes the influence of abnormal points on normal points. The frequency reconstructor transforms the input sequence into the frequency domain using a Fourier transform and extracts the relationship between frequencies to reconstruct anomalous frequency bands. The dual-view adversarial learning mechanism aims to maximize the normal values in the reconstructed sequences and highlight anomalies and aid in their localization within the data. Through dual-view adversarial learning, ATF-UAD minimizes reconstructed value errors and maximizes the identification of residual outliers. We conducted extensive experiments on nine datasets from different domains, and ATF-UAD showed an average improvement of 6.94% in terms of F1 score compared to the state-of-the-art method.

Abnormality of mussel in the early developmental stages induced by graphene and triphenyl phosphate: In silico toxicogenomic data-mining, in vivo, and toxicity pathway-oriented approach.

Increasing number of complex mixtures of organic pollutants in coastal area (especially for nanomaterials and micro/nanoplastics associated chemicals) threaten aquatic ecosystems and their joint hazards are complex and demanding tasks. Mussels are the most sensitive marine faunal groups in the world, and their early developmental stages (embryo and larvae) are particularly susceptible to environmental contaminants, which can distinguish the probable mechanisms of mixture-induced growth toxicity. In this study, the potential critical target and biological processes affected by graphene and triphenyl phosphate (TPP) were developed by mining public toxicogenomic data. And their combined toxic effects were verified by toxicological assay at early developmental stages in filter-feeding mussels (embryo and larvae). It showed that interactions among graphene/TPP with 111 genes (ABCB1, TP53, SOD, CAT, HSP, etc.) affected phenotypes along conceptual framework linking these chemicals to developmental abnormality endpoints. The PPAR signaling pathway, monocarboxylic acid metabolic process, regulation of lipid metabolic process, response to oxidative stress, and gonad development were noted as the key molecular pathways that contributed to the developmental abnormality. Enriched phenotype analysis revealed biological processes (cell proliferation, cell apoptosis, inflammatory response, response to oxidative stress, and lipid metabolism) affected by the investigated mixture. Combined, our results supported that adverse effects induced by contaminants/ mixture could not only be mediated by single receptor signaling or be predicted by the simple additive effect of contaminants. The results offer a framework for better comprehending the developmental toxicity of environmental contaminants in mussels and other invertebrate species, which have considerable potential for hazard assessment of coastal mixture.

FOSS-Based Method for Thin-Walled Structure Deformation Perception and Shape Reconstruction.

To improve the accuracy of deformation perception and shape reconstruction of flexible thin-walled structures, this paper proposes a method based on the combination of FOSS (fiber optic sensor system) and machine learning. In this method, the sample collection of strain measurement and deformation change at each measuring point of the flexible thin-walled structure was completed by ANSYS finite element analysis. The outliers were removed by the OCSVM (one-class support vector machine) model, and the unique mapping relationship between the strain value and the deformation variables (three directions of x-, y-, and z-axis) at each point was completed by a neural-network model. The test results show that the maximum error of the measuring point in the direction of the three coordinate axes: the x-axis is 2.01%, the y-axis is 29.49%, and the z-axis is 15.52%. The error of the coordinates in the y and z directions was large, and the deformation variables were small, the reconstructed shape had good consistency with the deformation state of the specimen under the existing test environment. This method provides a new idea with high accuracy for real-time monitoring and shape reconstruction of flexible thin-walled structures such as wings, helicopter blades, and solar panels.

Coupling strength of human-natural systems mediates the response of ecosystem services to land use change.

Addressing the dynamics of human-natural systems (HNS) driven by land use change (LC) is a key challenge for the sustainable development of ecosystem services (ES). However, how changes to the HNS coupling relationships affect ES is rarely reported. We used network analysis methods to construct an HNS correlation network in the Loess Plateau based on the correlation between the main components of HNS, such as ES, human factors, landscape pattern, vegetation cover, climate change and geomorphic characteristics, and quantitatively described the HNS coupling relationships through key network attributes. We analyzed the variation in HNS network attributes and their relationships with ES along an LC intensity gradient. The results show that carbon storage and soil conservation in the Loess Plateau increased by 0.56% and 0.26%, respectively, during the study period, while the habitat quality and water yield decreased by 0.11% and 0.18%, respectively. An increase in LC intensity reduces connectivity and density in the HNS network, which results in looser connections among HNS components. Importantly, we found that HNS network attributes explained 85% of ES variation across different LC intensity gradients and that connectivity and density had the strongest explanatory power. This means that LC mainly affects ES dynamics by changing the coupling strength of HNS. Our research offers a new perspective for linking LC-HNS-ES, which will help guide practitioners toward establishing and maintaining the sustainability of human well-being amidst changing HNS.

Development and validation of a urinary microRNA biomarker panel as a tool for early detection of prostate cancer in a Chinese population.

INTRODUCTION: Urinary microRNAs (miRNAs) may serve as promising biomarkers for non-invasive early detection of prostate cancer (PCa). We aimed to identify multi-miRNA urinary biomarker panel for early detection of PCa. METHODS: Urine samples from 83 PCa patients and 88 healthy control subjects in a Chinese population were collected for miRNA profiling. The absolute expression of 360 unique miRNAs were measured in each sample using a highly sensitive and robust RT-qPCR workflow. Candidate urinary miRNA biomarkers were identified based on differential expression between PCa patients and healthy controls. Multi-miRNA biomarker panels were optimised for detection of PCa using three regression algorithms (Lasso, Stepwise, Exhaustive) to identify an optimal biomarker panel with best detection performance and least number of miRNAs. RESULTS: A total of 312 miRNAs were detected in urine samples, 10 candidate urinary miRNA biomarkers differentially expressed between PCa and healthy samples were identified. A panel comprising these 10 miRNAs detected PCa with an area under the curve (AUC) of 0.738. Optimization of multi-miRNA panels resulted in a 6-miRNA biomarker panel (hsa-miR-375, hsa-miR-520d-5p, hsa-miR-199b-5p, hsa-miR-518e-5p, hsa-miR-31-3p and hsa-miR-4306) that had an AUC of 0.750. CONCLUSION: We identified a urinary miRNA biomarker panel for early detection of PCa in a Chinese population.

Combined toxic effects of nanoplastics and norfloxacin on mussel: Leveraging biochemical parameters and gut microbiota.

Antibiotics and nanoplastics (NPs) are among the two most concerned and studied marine emerging contaminants in recent years. Given the large number of different types of antibiotics and NPs, there is a need to apply efficient tools to evaluate their combined toxic effects. Using the thick-shelled mussel (Mytilus coruscus) as a marine ecotoxicological model, we applied a battery of fast enzymatic activity assays and 16S rRNA sequencing to investigate the biochemical and gut microbial response of mussels exposed to antibiotic norfloxacin (NOR) and NPs (80 nm polystyrene beads) alone and in combination at environmentally relevant concentrations. After 15 days of exposure, NPs alone significantly inhibited superoxide dismutase (SOD) and amylase (AMS) activities, while catalase (CAT) was affected by both NOR and NPs. The changes in lysozyme (LZM) and lipase (LPS) were increased over time during the treatments. Co-exposure to NPs and NOR significantly affected glutathione (GSH) and trypsin (Typ), which might be explained by the increased bioavailable NOR carried by NPs. The richness and diversity of the gut microbiota of mussels were both decreased by exposures to NOR and NPs, and the top functions of gut microbiota that were affected by the exposures were predicted. The data fast generated by enzymatic test and 16S sequencing allowed further variance and correlation analysis to understand the plausible driving factors and toxicity mechanisms. Despite the toxic effects of only one type of antibiotics and NPs being evaluated, the validated assays on mussels are readily applicable to other antibiotics, NPs, and their mixture.

Characterization of oxidative damage induced by nanoparticles via mechanism-driven machine learning approaches.

The wide application of TiO2-based engineered nanoparticles (nTiO2) inevitably led to release into aquatic ecosystems. Importantly, increasing studies have emphasized the high risks of nTiO2 to coastal environments. Bivalves, the representative benthic filter feeders in coastal zones, acted as important roles to assess and monitor the toxic effects of nanoparticles. Oxidative damage was one of the main toxic mechanisms of nTiO2 on bivalves, but the experimental variables/nanomaterial characteristics were diverse and the toxicity mechanism was complex. Therefore, it was very necessary to develop machine learning model to characterize and predict the potential toxicity. In this study, thirty-six machine learning models were built by nanodescriptors combined with six machine learning algorithms. Among them, random forest (RF) - catalase (CAT), k-neighbors classifier (KNN) - glutathione peroxidase (GPx), neural networks - multilayer perceptron (ANN) - glutathione s-transferase (GST), random forest (RF) - malondialdehyde (MDA), random forest (RF) - reactive oxygen species (ROS), and extreme gradient boosting decision tree (XGB) - superoxide dismutase (SOD) models performed good with high accuracy and balanced accuracy for both training sets and external validation sets. Furthermore, the best model revealed the predominant factors (exposure concentration, exposure periods, and exposure matrix) influencing the oxidative stress induced by nTiO2. These results showed that high exposure concentrations and short exposure-intervals tended to cause oxidative damage to bivalves. In addition, gills and digestive glands could be vulnerable to nTiO2-induced oxidative damage as tissues/organs differences were the important factors controlling MDA activity. This study provided insights into important nano-features responsible for the different indicators of oxidative stress and thereby extended the application of machine learning approaches in toxicological assessment for nanoparticles.

Toxicological mechanism of cadmium in the clam Ruditapes philippinarum using combined ionomic, metabolomic and transcriptomic analyses.

Cadmium (Cd) contamination in marine environment poses great risks to the organisms due to its potential adverse effects. In the present study, the toxicological effects and mechanisms of Cd at environmentally relevant concentrations (5 and 50 µg/L) on clam Ruditapes philippinarum after 21 days were investigated by combined ionomic, metabolomic, and transcriptomic analyses. Results showed that the uptake of Cd significantly decreased the concentrations of Cu, Zn, Sr, Se, and Mo in the whole soft tissue from 50 µg/L Cd-treated clams. Significantly negative correlations were observed between Cd and essential elements (Zn, Sr, Se, and Mo). Altered essential elements homeostasis was associated with the gene regulation of transport and detoxification, including ATP-binding cassette protein subfamily B member 1 (ABCB1) and metallothioneins (MT). The crucial contribution of Se to Cd detoxification was also found in clams. Additionally, gene set enrichment analysis showed that Cd could interfere with proteolysis by peptidases and decrease the translation efficiency at 50 µg/L. Cd inhibited lipid metabolism in clams and increased energy demand by up-regulating glycolysis and TCA cycle. Osmotic pressure was regulated by free amino acids, including alanine, glutamate, taurine, and homarine. Meanwhile, significant alterations of some differentially expressed genes, such as dopamine-ß-hydroxylase (DBH), neuroligin (NLGN), NOTCH 1, and chondroitin sulfate proteoglycan 1 (CSPG1) were observed in clams, which implied potential interference with synaptic transmission. Overall, through integrating multiple omics, this study provided new insights into the toxicological mechanisms of Cd, particularly in those mediated by dysregulation of essential element homeostasis.

Bioaccumulation and human health implications of trace metals in oysters from coastal areas of China.

This study recompiled a national dataset to characterize the pollution level and health risk of cadmium (Cd), copper (Cu), lead (Pb) and zinc (Zn) in oysters along the coastal areas of China. Results showed that the median concentrations of Cd, Cu, Pb and Zn in nationwide oysters were 5.5, 335, 1.3 and 1280 mg/kg dry weight, respectively. Generally, oysters from the north coasts presented lower metal pollution and higher quality than those from the south. The regional characteristics of trace metals in oysters might be contributed by the interspecific differences. Nationally, the noncarcinogenic risk posed by these four metals in oysters was relatively low, with the risk only occurring in a few hotspots such as the Pearl River Estuary and the Jiulong River Estuary. However, more attention should be paid to the carcinogenic risk of Cd, and priority should be given to formulating control measures to mitigate Cd pollution.

Throughput Maximization Using Deep Complex Networks for Industrial Internet of Things.

The high-density Industrial Internet of Things needs to meet the requirements of high-density device access and massive data transmission, which requires the support of multiple-input multiple-output (MIMO) antenna cognitive systems to keep high throughput. In such a system, spectral efficiency (SE) optimization based on dynamic power allocation is an effective way to enhance the network throughput as the channel quality variations significantly affect the spectral efficiency performance. Deep learning methods have illustrated the ability to efficiently solve the non-convexity of resource allocation problems induced by the channel multi-path and inter-user interference effects. However, current real-valued deep-learning-based power allocation methods have failed to utilize the representational capacity of complex-valued data as they regard the complex-valued channel data as two parts: real and imaginary data. In this paper, we propose a complex-valued power allocation network (AttCVNN) with cross-channel and in-channel attention mechanisms to improve the model performance where the former considers the relationship between cognitive users and the primary user, i.e., inter-network users, while the latter focuses on the relationship among cognitive users, i.e., intra-network users. Comparison experiments indicate that the proposed AttCVNN notably outperforms both the equal power allocation method (EPM) and the real-valued and the complex-valued fully connected network (FNN, CVFNN) and shows a better convergence rate in the training phase than the real-valued convolutional neural network (AttCNN).

Beyond the exposure phase: Microplastic depuration and experimental implications.

Currently, most studies focus on the effect of microplastics (MPs) in the exposure phase, but pay limited attention to the depuration phase. Depuration is a promising practice to achieve safe aquaculture production, which is also helpful to understand the long-term impact of MPs. Therefore, investigating the post-exposure scenarios of MPs has great practical significance. In order to provide implications for future research, this work attempted to systematize the current findings and knowledge gaps regarding the depuration of MPs. More specifically, three methods, including direct fitting, one-compartment kinetic model and interval observation, for estimating the retention time of MPs to further determine the minimum depuration time were introduced, in which the one-compartment kinetic model could also be used to calculate the depuration rate constant and biological half-life of MPs. Moreover, the post-exposure effect of MPs generally presented three scenarios: incomplete reversal (legacy effect), return to control level (recovery) and stimulatory response (hormesis-like effect). In addition, the possible tissue translocation of MPs, the influence of food abundance and body shape on MPs egestion, and the potential interaction with environmental factors, have aroused great scientific concerns and need further exploration and clarification.

Reconciling the actual and nominal exposure concentrations of microplastics in aqueous phase: Implications for risk assessment and deviation control.

The deviation between actual and nominal concentrations of microplastics (MPs), as a long-standing issue, has been critically commented. However, there is still a lack of quantitative assessment and reconciling practice on the deviation. In this study, a total of 210 deviations were recompiled to thoroughly examine this issue. It was shown that up to 81 (39%) deviations exceeded the recommended ± 20% variation specification, highlighting that the deviation of MPs should not be neglected. This study attempted to reconcile the deviation based on the most prominent driving factors. Specifically, the game theory-based SHapley Additive exPlanations (SHAP) algorithm identified that the particle size was the most important factor affecting the deviation. Subsequently, at each size magnitude, a significant linear correlation between the logarithmic actual and nominal concentrations was determined, which provided a sound basis for estimating the actual concentration from the nominal one. Furthermore, deviations of different size classes were simulated through 10, 000 points, suggesting that the ± 20% deviation variation could be well maintained within a specific concentration range. Moreover, the potential interaction effects between factors were quantified by SHAP interaction values, with more detailed conversion bases proposed. Additionally, several control measures were recommended to reduce the deviation of MPs.