Chemical Research in Toxicology最新文献

{"title":"Ultrasensitive and Specific Detection of 8-Oxoguanine DNA Glycosylase via Locus-Specific Rolling Circle Amplification","authors":"Zi-Qin Huang,&nbsp;Jia-Hui Dong,&nbsp;Run-Hong Zhang,&nbsp;Ling-Li Zhao,&nbsp;Ying Liu,&nbsp;Ying-Lin Zhou* and Xin-Xiang Zhang*,&nbsp;","doi":"10.1021/acs.chemrestox.6c00172","DOIUrl":"10.1021/acs.chemrestox.6c00172","url":null,"abstract":"<p >8-Oxoguanine DNA glycosylase 1 (OGG1) is a key enzyme for maintaining genomic integrity, as it specifically recognizes and excises 8-oxoguanine (OG), a major oxidative DNA lesion, to initiate base excision repair. Dysregulation of OGG1 activity is closely associated with genomic instability, apoptosis, and tumorigenesis. However, conventional methods for detecting OGG1 activity often lack the sensitivity required for trace-level analysis, limiting their application in early diagnostics and mechanistic studies. In this study, we report an innovative assay termed OG-specific rolling circle amplification (OG-RCA), which for the first time integrates OGG1-mediated OG excision with RCA and subsequent G-triplex formation for signal readout. This novel approach establishes a unique conversion strategy that translates enzymatic activity into quantifiable amplification signals, significantly enhancing detection sensitivity and specificity. After systematic optimization of key reaction conditions, including dsOG substrate concentration, enzyme amounts, and DNA probe concentrations, the OG-RCA assay achieved an exceptionally low limit of detection (LOD) of 3 × 10^–8 mg/mL (equivalent to ∼1.6 × 10^–5 U/mL) for OGG1 activity, surpassing existing methods. The assay also exhibited high specificity, showing minimal cross-reactivity with other DNA repair glycosylases. Moreover, spike-recovery experiments using HeLa cell protein extracts and 293T cell lysates confirmed its robustness in complex biological samples. The OG-RCA method not only provides a powerful tool for the ultrasensitive detection of DNA base damage biomarkers but also offers a novel platform for investigating DNA damage and repair mechanisms. It holds significant promise for applications involving limited biological samples and mechanistic studies of DNA repair.</p>","PeriodicalId":31,"journal":{"name":"Chemical Research in Toxicology","volume":"39 6","pages":"1246–1253"},"PeriodicalIF":3.8,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148141918","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sex-Specific Modulation of Drug Metabolizing Enzymes, Transporters, and Pro-Inflammatory Cytokines by Arsenic Trioxide in C57Bl/6 Mice","authors":"Sara R. El-Mahrouk,&nbsp; and ,&nbsp;Ayman O. S. El-Kadi*,&nbsp;","doi":"10.1021/acs.chemrestox.6c00114","DOIUrl":"10.1021/acs.chemrestox.6c00114","url":null,"abstract":"<p >Arsenic trioxide (ATO) is both a life-saving therapy for acute promyelocytic leukemia and a systemic toxicant whose hepatic effects remain incompletely defined. This study examined how a single clinically relevant ATO dose (8 mg/kg, i.p.) acutely remodels hepatic xenobiotic-metabolizing enzymes, arsenic transporters, and pro-inflammatory mediators in male and female C57Bl/6 mice. Mice were treated with ATO or saline and livers were collected at 6 and 24 h for integrated mRNA and protein profiling of major Cytochrome P450 (CYP) families, aquaglyceroporins (<i>Aqp3/7/9</i>), ATP-binding cassette (<i>Abcb1, Abcc1–6</i>) transporters, and cytokines (<i>Tnf-α, Il-1β, Il-6</i>). ATO induced highly sex-, time-, and isoform-specific reprogramming. Females exhibited a wider and earlier decline in several female-predominant CYP2, CYP3, and CYP4 isoforms, including a more pronounced reduction in hepatic CYP3A, CYP4A, and CYP4F protein abundance. In contrast, males showed mainly transcriptional induction of specific genes (<i>Cyp1a1</i>, <i>Cyp2a</i>, <i>Cyp3a13</i>, and <i>Cyp4a</i>), accompanied by comparatively modest decreases in overall CYP protein levels. <i>Aqp</i> and <i>Abc</i> transporters were differentially modulated, with males displaying early, relatively monotonic upregulation of <i>Abcb1/Abcc</i> efflux systems, while females exhibited higher basal <i>Abc</i> expression but more complex, biphasic regulation of both influx (<i>Aqp7/9</i>) and efflux pathways. These transcriptional changes paralleled a transient inflammatory response, including early <i>Tnf-α</i> induction and female-specific <i>Il-6</i> elevation. Collectively, these findings highlight sex-dependent modulation of hepatic ATO handling and drug metabolizing capacity, with important implications for risk assessment and individualized ATO containing regimens.</p>","PeriodicalId":31,"journal":{"name":"Chemical Research in Toxicology","volume":"39 6","pages":"1218–1233"},"PeriodicalIF":3.8,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147830543","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Silver-Form-Driven Nanotoxicity Reveals Spirodela intermedia W. Koch as a Potential Bioindicator for Aquatic Systems","authors":"Maia Sabrina Cruz,&nbsp;Mailén Petri,&nbsp;Luca Terraciano,&nbsp;María Gabriela Lagorio,&nbsp;Graciela Alicia González* and Virginia Diz*,&nbsp;","doi":"10.1021/acs.chemrestox.5c00490","DOIUrl":"10.1021/acs.chemrestox.5c00490","url":null,"abstract":"<p >Silver nanoparticles (AgNPs) and silver sulfide nanoparticles (Ag₂SNPs) are increasingly being used in diverse industries due to their unique physicochemical properties. However, their release into aquatic environments raises concerns regarding their potential ecological impact, making them an excellent model system for evaluating potential biomonitors. In this study, we evaluated and compared the phytotoxic effects of AgNPs and Ag₂SNPs suspensions on the free-floating macrophyte <i>Spirodela intermedia</i> W. Koch, a species with broad distribution from Mexico to South America and promising potential as a bioindicator. The designed protocol involved both types of nanoparticles, which were synthesized and characterized via UV–vis spectrophotometry and other techniques. In addition, morphometric parameters (frond number, frond abscission, and chlorosis), as well as enzymatic activity of ascorbate oxidase (AO) and chlorophyll-a variable fluorescence, were assessed after exposure to environmentally relevant concentrations. Results indicated that AgNPs induced significantly higher toxic effects than Ag₂SNPs, with marked frond abscission, chlorosis, AO activity reduction, and damage to the photosynthetic apparatus. These findings indicate that, as expected, the chemical form of silver nanoparticles plays a key role in phytotoxicity and that <i>S. intermedia</i> exhibits adequate sensitivity to be proposed as a bioindicator candidate in aquatic ecosystems, thereby expanding the global portfolio of bioindicator species and highlighting the ecological value of native flora for sustainable water quality monitoring.</p>","PeriodicalId":31,"journal":{"name":"Chemical Research in Toxicology","volume":"39 6","pages":"1102–1110"},"PeriodicalIF":3.8,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148050928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Endogenous DNA Damage and Its Role in Human Disease","authors":"Eva-Maria Bryan*,&nbsp; and ,&nbsp;Clare Lauderback,&nbsp;","doi":"10.1021/acs.chemrestox.5c00539","DOIUrl":"10.1021/acs.chemrestox.5c00539","url":null,"abstract":"<p >Endogenous DNA damage from metabolism and genome dynamics drives mutation, instability, and disease. New insights from ACS Fall 2025 reveal chemical diversity, metabolic drivers, and repair limits that shape cancer risk and biomarker development.</p>","PeriodicalId":31,"journal":{"name":"Chemical Research in Toxicology","volume":"39 6","pages":"1055–1058"},"PeriodicalIF":3.8,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/pdf/10.1021/acs.chemrestox.5c00539","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146217950","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Role of HGF/c-Met/PI3K/AKT Pathway Regulated by α7-nAChR/β2-AR in Nicotine-Promoted Malignant Progression of Nonsmall Cell Lung Cancer A549 Cells","authors":"Zihan Li,&nbsp;Boxuan Xie,&nbsp;Xilei Liu,&nbsp;Zihan He,&nbsp;Zhongwei Zhang,&nbsp;Jianming Zhou,&nbsp;Tong Zhou and Huai Wang*,&nbsp;","doi":"10.1021/acs.chemrestox.5c00464","DOIUrl":"10.1021/acs.chemrestox.5c00464","url":null,"abstract":"<p >As the most frequently occurring cancer worldwide, lung cancer is notoriously diagnosed at advanced stages, resulting in high mortality rates. The primary factor underlying this persistent global health burden remains tobacco consumption. Nicotine, a key component of cigarette smoke, is one of the major contributing factors to the development of lung cancer, but the molecular mechanisms remain incompletely elucidated. α7-nAChR, β2-AR, and HGF/c-Met are known to contribute to lung cancer development, respectively. However, the specific signaling cascade through which they interact in nicotine-induced malignancy is unclear. In this study, using CCK-8 assays across a range of concentrations (0–5 μM), we determined that 1 μM nicotine treatment for 48 h optimally enhances A549 cell proliferation, establishing this condition for subsequent experiments. This study uncovers nicotine’s role in driving malignant behaviors including proliferation, migration, and invasion in nonsmall cell lung cancer A549 cells. Notably, nicotine potently stimulated migration and invasion, accompanied by upregulation of Cyclin D1 and MMP-2, and downregulation of BAX, BAD, and Caspase-3. Mechanistically, nicotine induced synergistic engagement of α7-nAChR and β2-AR, leading to activation of the HGF/c-Met/PI3K/AKT axis and enhancing the secretion of both HGF and MMP-2. Importantly, we reveal a previously unrecognized bidirectional regulatory loop between α7-nAChR and β2-AR that functionally converges on the HGF/c-Met axis, which acts as a critical signaling hub to drive the downstream PI3K/AKT pathway and facilitate tumor progression. Our findings provide the first evidence of a coordinated α7-nAChR/β2-AR interface regulating HGF/c-Met signaling that orchestrates both intracellular signaling and extracellular secretory programs in nicotine-promoted lung cancer progression. This offers innovative insights for identifying potential antitumor therapeutic targets and presents novel perspectives for the prevention and clinical management of lung cancer, particularly in smokers with nonsmall cell lung cancer.</p>","PeriodicalId":31,"journal":{"name":"Chemical Research in Toxicology","volume":"39 6","pages":"1088–1101"},"PeriodicalIF":3.8,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147979461","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PFDA Influences the Structure of the Globular Milk Protein, α-Lactalbumin: Implications for Its Ability to Bind a Physiologically Crucial Cofactor","authors":"Daisy L. Wilson,&nbsp;Randhal S. Ramirez Orozco,&nbsp;Lela Vukovic* and Mahesh Narayan*,&nbsp;","doi":"10.1021/acs.chemrestox.5c00547","DOIUrl":"10.1021/acs.chemrestox.5c00547","url":null,"abstract":"<p >Perfluorodecanoic acid (PFDA) ingestion is associated with liver, immune, developmental, and reproductive effects. Nevertheless, the molecular mechanisms underlying PFDA toxicity remain poorly understood. For example, despite its established presence in human milk, cow milk, and infant formula, its molecular interactions with constituent proteins and their consequences require further study. Here, we report the outcomes associated with the interaction between PFDA and α-lactalbumin (ALAC), a calcium-binding whey protein essential for nutrition and lactose production. Absorbance and Trp fluorescence data reveal PFDA-dependent changes consistent with interactions that perturb the native disposition of the optically active chromophores. Deconvolution of the amide I region of PFDA: protein IR spectra suggest dose-dependent distortions of PFDA in helical and sheet topologies. Ca²⁺-binding kinetics suggest that the “forever” chemical compromised metal-ion binding to the protein in a dose-dependent manner, reflecting impaired metal-dependent structural stabilization from the molten-globule-like apo-state to the native and biologically active holo-state. Molecular dynamics simulations identified two preferential PFDA binding regions enriched in hydrophobic and positively charged residues and showed that local rearrangements in ALAC’s unstructured N-terminus coils can generate tightly bound PFDA states with favorable interaction energies. Combined, these results reveal a coherent molecular mechanism in which PFDA anchors to the ALAC surface, disrupts secondary structure organization, and weakens Ca²⁺ binding. Considering ALAC’s role in early infant nutrition and human health, these findings provide a mechanistic insight into how PFAS exposure may compromise protein function in the postnatal environment.</p>","PeriodicalId":31,"journal":{"name":"Chemical Research in Toxicology","volume":"39 6","pages":"1111–1122"},"PeriodicalIF":3.8,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147687086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Few-Shot Prediction of Toxicity of Ionic Liquids Supported by Attentive Model-Agnostic Meta-Learning","authors":"Karol Baran*,&nbsp;Tomasz Derron,&nbsp;Joachim Eichenlaub and Adam Kloskowski,&nbsp;","doi":"10.1021/acs.chemrestox.6c00022","DOIUrl":"10.1021/acs.chemrestox.6c00022","url":null,"abstract":"<p >Prediction of chemical compounds’ toxicity enables efficient and rapid screening at the cost of utilizing experimental data as a foundation for artificial intelligence (AI) models. Given the constraints of limited data availability, few-shot learning techniques, like those based on meta-learning models, may be beneficial. In this study, the ionic liquid (IL) toxicity dataset was employed. Initially, models underwent pretraining on data from related domains before being adapted in a few-shot manner, adhering to the Model-Agnostic Meta-Learning (MAML) algorithm. The MAML algorithm utilized a neural network to predict toxicity based on the descriptors of the molecular structure. MAML enhanced the accuracy of IL toxicity prediction even with limited data, particularly when similar tasks were provided during meta-training, such as between <i>Escherichia coli</i> and <i>Vibrio fischeri</i>. Its effectiveness, however, is dependent on the diversity of the chemical space of the task for which adaptation is performed. As the final part of the study, a new version of MAML (i.e., attentive MAML) was proposed to address the limitations of the basic method. In our Att-MAML modified version of the algorithm, a part of the neural network was dedicated to adapting the importance of latent features with respect to the given test task. This approach allowed a significant reduction in standard deviation and overall improvement of the performance metrics in the low-data regime. This work advances the toxicity prediction in extremely low-data regimes beyond the current state of the art. Incorporation of attentive latent feature scaling (with Att-MAML) handles the limitations inherent to extremely low-data regimes. This novel method allows reductions in the standard deviation of metric values even while adapting on sets as small as four records.</p>","PeriodicalId":31,"journal":{"name":"Chemical Research in Toxicology","volume":"39 6","pages":"1123–1140"},"PeriodicalIF":3.8,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147939030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Classification of Thyroid Peroxidase (TPO) Inhibitors Using Transfer Learning with SMILES Embeddings","authors":"Geven Piir,&nbsp;Sulev Sild,&nbsp;Eliana Spilioti,&nbsp;Dimitra Nikolopoulou,&nbsp;Effrosyni Katsanou,&nbsp;Ingrid Langezaal and Uko Maran*,&nbsp;","doi":"10.1021/acs.chemrestox.6c00058","DOIUrl":"10.1021/acs.chemrestox.6c00058","url":null,"abstract":"<p >Thyroid hormones (THs) regulate many processes in mammals and, therefore, affect every organ in the body. Thyroid peroxidase (TPO) is an essential enzyme for the successful biosynthesis of THs. Although TPO inhibition is a well-documented molecular initiating event (MIE) in thyroid hormone system disruption adverse outcome pathways (AOPs), experimental methods and computational models to assess TPO activity are lacking. Efficient computational new approach methodologies (NAMs) are a viable solution for identifying TPO inhibitors from a large pool of agrochemicals. The aim of this study was to investigate the suitability of SMILES embeddings generated using a specialized language model (SLM) based on a pretrained deep neural network (DNN) for applying a transfer learning approach in the development of quantitative structure–activity relationships for classifying TPO inhibitors. Traditional theoretical molecular descriptors were used for comparison. Two different molecular descriptor sets resulted in Random Forest (RF) models that performed similarly on the training and test sets, while the sensitivity for the external validation set was substantially different between the two models (0.788 vs 0.490). Comparison of the predictions with the TPO inhibition data of the chemicals assessed by EFSA and EU-NETVAL laboratories showed good agreement. At the same time, analysis of experimental data from other sources showed some conflicting estimates. This suggests that further and more precise studies are needed for some compounds. This study advances in silico methodologies by implementing transfer learning for QSAR modeling from text representations (e.g., SMILES) using the pretrained Bidirectional Encoder Representations from Transformers (BERT) architecture. While traditional QSAR approach relies on molecular descriptors, this evaluation shows that model-generated SMILES embeddings can expand the applicability domain, indicating a more robust representation of structural information compared to traditional molecular descriptors.</p>","PeriodicalId":31,"journal":{"name":"Chemical Research in Toxicology","volume":"39 6","pages":"1166–1177"},"PeriodicalIF":3.8,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/pdf/10.1021/acs.chemrestox.6c00058","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148141886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hemin Recapitulates the Labile Iron Pool in the Cellular Fenton Reaction with DNA","authors":"Aaron M. Fleming*,&nbsp; and ,&nbsp;Cynthia J. Burrows*,&nbsp;","doi":"10.1021/acs.chemrestox.6c00203","DOIUrl":"10.1021/acs.chemrestox.6c00203","url":null,"abstract":"<p >The iron-Fenton reaction in biology is influenced by diverse cellular molecules that coordinate with redox-active ferrous ion. Contemporary research proposed that hydroxyl radical or a ferryl species (HO^•/Fe=O²⁺) are the primary oxidants; however, this was challenged by the observation that physiological bicarbonate redirects the reaction to form carbonate radical anion (CO₃^•–). Questions remained regarding the roles of O₂ concentration, ascorbate, and iron speciation in CO₃^•– formation. Accordingly, in cellulo studies were conducted under physiological O₂ (∼25 μM) with ascorbate replenishment to monitor bicarbonate-dependent telomeric DNA damage. Under these conditions,. physiological bicarbonate (25 mM) yielded 2′-deoxyguanosine-specific oxidation consistent with CO₃^•– formation at a ratio exceeding 80:1 relative to HO^•/Fe=O²⁺. In parallel in vitro experiments using a plasmid nicking assay, the cellular low molecular weight (LMW) ultrafiltrate was used as the source of iron and its endogenous coordination partners; under physiological O₂, bicarbonate, and 500 nM H₂O₂ (the concentration required to produce detectable signal), the DNA damage profile was consistent with exclusive CO₃^•– formation, mirroring the cell culture result. A panel of iron complexes approximating the intracellular labile iron pool (5 μM) was examined: hexaaquo-ferrous ion, ferrous citrate, ferrous α-ketoglutarate, ferrous pyrophosphate, ferrous glutathione, and hemin. Of these, only hemin reproduced the bicarbonate-dependent CO₃^•– damage profile observed in cells with 100 nM H₂O₂ and 25 mM bicarbonate present. This finding was corroborated using a defined biomimetic metabolome in which hemin, ferrous ion, or their combination was tested; hemin consistently supported CO₃^•– as the dominant oxidant. Roles for the Udenfriend reaction (Fe(II), O₂, and reductant) and superoxide dismutase were also studied. Collectively, these results identify heme iron as a likely candidate to drive CO₃^•– formation via the bicarbonate iron-Fenton reaction to damage dG in DNA during endogenous oxidative stress.</p>","PeriodicalId":31,"journal":{"name":"Chemical Research in Toxicology","volume":"39 6","pages":"1254–1266"},"PeriodicalIF":3.8,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148147380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prediction of Ligand Binding to Transthyretin Using Machine Learning Algorithms and Low-Dimensional Molecular Descriptors: A Tox24 Challenge Study","authors":"Filip Stefaniak*,&nbsp;","doi":"10.1021/acs.chemrestox.6c00052","DOIUrl":"10.1021/acs.chemrestox.6c00052","url":null,"abstract":"<p >This study presents a comprehensive machine learning approach for predicting the percent displacement of ANSA from human transthyretin (TTR) at fixed assay conditions as defined in the Tox24 Challenge. TTR is a critical serum transport protein for thyroid hormones, and its disruption by environmental chemicals can lead to endocrine system dysregulation with serious developmental and metabolic consequences. However, the scarcity of large, chemically diverse data sets and the lack of standardized experimental protocols have limited the computational prediction of TTR binding, restricting the development of robust predictive models applicable to broad chemical spaces. The described pipeline uses computationally efficient, low-dimensional (0D–2D) molecular descriptors and fingerprints. This eliminates the need for costly 3D conformational analysis. Following the systematic benchmarking of individual machine learning (ML) methods, hyperparameter optimization was performed using Optuna for two gradient boosting algorithms: CatBoost and XGBoost. The feature importance analysis revealed complementary learning strategies between these algorithms. The proposed consensus model achieved an RMSE of 21.60 on the blind test set, ranking 15th among 79 participating teams. Chemical space analysis using PCA and t-SNE confirmed that, except for two outliers, the test compounds fell within the distribution for the training set. Postchallenge analyses evaluated the effect of the cross-validation strategy (random vs cluster-based split) and descriptor dimensionality (2D-only, 3D-only, or mixed) on model performance. To facilitate broader adoption, a freely accessible web server was developed, enabling rapid toxicity prediction across multiple Tox21 and Tox24 end points without requiring computational expertise (https://toxpred.genesilico.pl/). This work demonstrates that low-dimensional molecular descriptors combined with optimized consensus ML methods can achieve competitive predictive performance, making high-throughput toxicity screening practical for drug discovery, environmental risk assessment, and regulatory decision-making.</p>","PeriodicalId":31,"journal":{"name":"Chemical Research in Toxicology","volume":"39 6","pages":"1152–1165"},"PeriodicalIF":3.8,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148141910","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}