Molecular basis for recognition of Gly/N-degrons by CRL2ZYG11B and CRL2ZER1.

N-degron pathways are a set of proteolytic systems that target the N-terminal destabilizing residues of substrates for proteasomal degradation. Recently, the Gly/N-degron pathway has been identified as a new branch of the N-degron pathway. The N-terminal glycine degron (Gly/N-degron) is recognized by ZYG11B and ZER1, the substrate receptors of the Cullin 2-RING E3 ubiquitin ligase (CRL2). Here we present the crystal structures of ZYG11B and ZER1 bound to various Gly/N-degrons. The structures reveal that ZYG11B and ZER1 utilize their armadillo (ARM) repeats forming a deep and narrow cavity to engage mainly the first four residues of Gly/N-degrons. The α-amino group of the Gly/N-degron is accommodated in an acidic pocket by five conserved hydrogen bonds. These structures, together with biochemical studies, decipher the molecular basis for the specific recognition of the Gly/N-degron by ZYG11B and ZER1, providing key information for future structure-based chemical probe design.

A chemical probe to modulate human GID4 Pro/N-degron interactions.

The C-terminal to LisH (CTLH) complex is a ubiquitin ligase complex that recognizes substrates with Pro/N-degrons via its substrate receptor Glucose-Induced Degradation 4 (GID4), but its function and substrates in humans remain unclear. Here, we report PFI-7, a potent, selective and cell-active chemical probe that antagonizes Pro/N-degron binding to human GID4. Use of PFI-7 in proximity-dependent biotinylation and quantitative proteomics enabled the identification of GID4 interactors and GID4-regulated proteins. GID4 interactors are enriched for nucleolar proteins, including the Pro/N-degron-containing RNA helicases DDX21 and DDX50. We also identified a distinct subset of proteins whose cellular levels are regulated by GID4 including HMGCS1, a Pro/N-degron-containing metabolic enzyme. These data reveal human GID4 Pro/N-degron targets regulated through a combination of degradative and nondegradative functions. Going forward, PFI-7 will be a valuable research tool for investigating CTLH complex biology and facilitating development of targeted protein degradation strategies that highjack CTLH E3 ligase activity.

C-terminal glutamine acts as a C-degron targeted by E3 ubiquitin ligase TRIM7.

The exposed N-terminal or C-terminal residues of proteins can act, in cognate sequence contexts, as degradation signals (degrons) that are targeted by specific E3 ubiquitin ligases for proteasome-dependent degradation by N-degron or C-degron pathways. Here, we discovered a distinct C-degron pathway, termed the Gln/C-degron pathway, in which the B30.2 domain of E3 ubiquitin ligase TRIM7 (TRIM7B30.2) mediates the recognition of proteins bearing a C-terminal glutamine. By determining crystal structures of TRIM7B30.2 in complexes with various peptides, we show that TRIM7B30.2 forms a positively charged binding pocket to engage the "U"-shaped Gln/C-degron. The four C-terminal residues of a substrate play an important role in C-degron recognition, with C-terminal glutamine as the principal determinant. In vitro biochemical and cellular experiments were used to further analyze the substrate specificity and selective degradation of the Gln/C-degron by TRIM7.

Crystal Structure Assignment for Unknown Compounds from X-ray Diffraction Patterns with Deep Learning.

Determining the structures of previously unseen compounds from experimental characterizations is a crucial part of materials science. It requires a step of searching for the structure type that conforms to the lattice of the unknown compound, which enables the pattern matching process for characterization data, such as X-ray diffraction (XRD) patterns. However, this procedure typically places a high demand on domain expertise, thus creating an obstacle for computer-driven automation. Here, we address this challenge by leveraging a deep-learning model composed of a union of convolutional residual neural networks. The accuracy of the model is demonstrated on a dataset of over 60,000 different compounds for 100 structure types, and additional categories can be integrated without the need to retrain the existing networks. We also unravel the operation of the deep-learning black box and highlight the way in which the resemblance between the unknown compound and a structure type is quantified based on both local and global characteristics in XRD patterns. This computational tool opens new avenues for automating structure analysis on materials unearthed in high-throughput experimentation.

EGLN3 attenuates gastric cancer cell malignant characteristics by inhibiting JMJD8/NF-κB signalling activation independent of hydroxylase activity.

BACKGROUND: The expression of Egl-9 family hypoxia-inducible factor 3 (EGLN3) is notably decreased in various malignancies, including gastric cancer (GC). While the predominant focus has been on the hydroxylase activity of EGLN3 for its antitumour effects, recent findings have suggested nonenzymatic roles for EGLN3. METHODS: This study assessed the clinical significance of EGLN3 expression in GC and explored the connection between EGLN3 DNA promoter methylation and transcriptional silencing. To investigate the effect of EGLN3 on GC cells, a gain-of-function strategy was adopted. RNA sequencing was conducted to identify the key effector molecules and signalling pathways associated with EGLN3. RESULTS: EGLN3 expression was significantly reduced in GC tissues, correlating with poorer patient prognosis. EGLN3 hypermethylation disrupts transcriptional equilibrium, contributing to deeper tumour invasion and lymph node metastasis, thus exacerbating GC progression. Conversely, restoration of EGLN3 expression in GC cells substantially inhibited cell proliferation and metastasis. EGLN3 was also found to impede the malignant progression of GC cells by downregulating Jumonji C domain-containing protein 8-mediated activation of the NF-κB pathway, independent of its hydroxylase activity. CONCLUSIONS: EGLN3 has the potential to hinder the spread of GC cells through a nonenzymatic mechanism, thereby shedding light on the complex nature of GC progression.

PdmIRD: missense variants pathogenicity prediction for inherited retinal diseases in a disease-specific manner.

Accurate discrimination of pathogenic and nonpathogenic variation remains an enormous challenge in clinical genetic testing of inherited retinal diseases (IRDs) patients. Computational methods for predicting variant pathogenicity are the main solutions for this dilemma. The majority of the state-of-the-art variant pathogenicity prediction tools disregard the differences in characteristics among different genes and treat all types of mutations equally. Since missense variants are the most common type of variation in the coding region of the human genome, we developed a novel missense mutation pathogenicity prediction tool, named Prediction of Deleterious Missense Mutation for IRDs (PdmIRD) in this study. PdmIRD was tailored for IRDs-related genes and constructed with the conditional random forest model. Population frequencies and a newly available prediction tool were incorporated into PdmIRD to improve the performance of the model. The evaluation of PdmIRD demonstrated its superior performance over nonspecific tools (areas under the curves, 0.984 and 0.910) and an existing eye abnormalities-specific tool (areas under the curves, 0.975 and 0.891). We also demonstrated the submodel that used a smaller gene panel further slightly improved performance. Our study provides evidence that a disease-specific model can enhance the prediction of missense mutation pathogenicity, especially when new and important features are considered. Additionally, this study provides guidance for exploring the characteristics and functions of the mutated proteins in a greater number of Mendelian disorders.

Antiferroelectricity-Induced Negative Thermal Expansion in Double Perovskite Pb2 CoMoO6.

Materials with negative thermal expansion (NTE) attract significant research attention owing to their unique physical properties and promising applications. Although ferroelectric phase transitions leading to NTE are widely investigated, information on antiferroelectricity-induced NTE remains limited. In this study, single-crystal and polycrystalline Pb2 CoMoO6 samples are prepared at high pressure and temperature conditions. The compound crystallizes into an antiferroelectric Pnma orthorhombic double perovskite structure at room temperature owing to the opposite displacements dominated by Pb2+ ions. With increasing temperature to 400 K, a structural phase transition to cubic Fm-3m paraelectric phase occurs, accompanied by a sharp volume contraction of 0.41%. This is the first report of an antiferroelectric-to-paraelectric transition-induced NTE in Pb2 CoMoO6 . Moreover, the compound also exhibits remarkable NTE with an average volumetric coefficient of thermal expansion αV = -1.33 × 10-5 K-1 in a wide temperature range of 30-420 K. The as-prepared Pb2 CoMoO6 thus serves as a prototype material system for studying antiferroelectricity-induced NTE.

Effects of Strong Inhibition of Cytochrome P450 3A and UDP glucuronosyltransferase 1A9 and Strong Induction of Cytochrome P450 3A on the Pharmacokinetics, Safety, and Tolerability of Soticlestat: Two Drug-Drug Interaction Studies in Healthy Volunteers.

Two open-label, phase 1 studies (NCT05064449, NCT05098041) investigated the effects of cytochrome P450 (CYP) 3A inhibition (via itraconazole), UDP glucuronosyltransferase (UGT) 1A9 inhibition (via mefenamic acid), and CYP3A induction (via rifampin) on the pharmacokinetics of soticlestat and its metabolites M-I and M3. In period 1 of both studies, participants received a single dose of soticlestat 300 mg. In period 2, participants received itraconazole on days 1-11 and soticlestat 300 mg on day 5 (itraconazole/mefenamic acid study; part 1); mefenamic acid on days 1-7 and soticlestat 300 mg on day 2 (itraconazole/mefenamic acid study; part 2); or rifampin on days 1-13 and soticlestat 300 mg on day 11 (rifampin study). Twenty-eight healthy adults participated in the itraconazole/mefenamic acid study (14 per part) and 15 participated in the rifampin study (mean age, 38.1-40.7 years; male, 79-93%). For maximum observed concentration, the geometric mean ratios (GMRs) of soticlestat + itraconazole, mefenamic acid, or rifampin to soticlestat alone were 116.6%, 107.3%, and 13.2%, respectively, for soticlestat; 10.7%, 118.0%, and 266.1%, respectively, for M-I, and 104.6%, 88.2%, and 66.6%, respectively, for M3. For area under the curve from time 0 to infinity, the corresponding GMRs were 124.0%, 100.6%, and 16.4% for soticlestat; 13.3%, 117.0%, and 180.8% for M-I; and 120.3%, 92.6%, and 58.4% for M3. Soticlestat can be administered with strong CYP3A and UGT1A9 inhibitors, but not strong CYP3A inducers (except for antiseizure medications, which will be further evaluated in ongoing phase 3 studies). In both studies, all treatment-emergent adverse events were mild or moderate. SIGNIFICANCE STATEMENT: These drug-drug interaction studies improve our understanding of the potential changes that may arise in soticlestat exposure in patients being treated with CYP3A inhibitors, UGT1A9 inhibitors, or CYP3A inducers. The results build on findings from previously published soticlestat studies and provide important information to help guide clinical practice. Soticlestat has shown positive phase 2 results and is currently in phase 3 development for the treatment of seizures in patients with Dravet syndrome and Lennox-Gastaut syndrome.

Sustainable management of tea wastes: resource recovery and conversion techniques.

As the demand for tea (Camellia sinensis) has grown across the world, the amount of biomass waste that has been produced during the harvesting process has also increased. Tea consumption was estimated at about 6.3 million tonnes in 2020 and is anticipated to reach 7.4 million tonnes by 2025. The generation of tea waste (TW) after use has also increased concurrently with rising tea consumption. TW includes clipped stems, wasted tea leaves, and buds. Many TW-derived products have proven benefits in various applications, including energy generation, energy storage, wastewater treatment, and pharmaceuticals. TW is widely used in environmental and energy-related applications. Energy recovery from low- and medium-calorific value fuels may be accomplished in a highly efficient manner using pyrolysis, anaerobic digestion, and gasification. TW-made biochar and activated carbon are also promising adsorbents for use in environmental applications. Another area where TW shows promise is in the synthesis of phytochemicals. This review offers an overview of the conversion procedures for TW into value-added products. Further, the improvements in their applications for energy generation, energy storage, removal of different contaminants, and extraction of phytochemicals have been reviewed. A comprehensive assessment of the sustainable use of TWs as environmentally acceptable renewable resources is compiled in this review.

Bioaccumulation and Trophodynamics of Novel Brominated Flame Retardants (NBFRs) in Marine Food Webs from the Arctic and Antarctic Regions.

The pervasive contamination of novel brominated flame retardants (NBFRs) in remote polar ecosystems has attracted great attention in recent research. However, understanding regarding the trophic transfer behavior of NBFRs in the Arctic and Antarctic marine food webs is limited. In this study, we examined the occurrence and trophodynamics of NBFRs in polar benthic marine sediment and food webs collected from areas around the Chinese Arctic Yellow River Station (n = 57) and Antarctic Great Wall Station (n = 94). ∑7NBFR concentrations were in the range of 1.27-7.47 ng/g lipid weight (lw) and 0.09-1.56 ng/g lw in the Arctic and Antarctic marine biota, respectively, among which decabromodiphenyl ethane (DBDPE) was the predominant compound in all sample types. The biota-sediment bioaccumulation factors (g total organic carbon/g lipid) of NBFRs in the Arctic (0.85-3.40) were 4-fold higher than those in the Antarctica (0.13-0.61). Trophic magnification factors (TMFs) and their 95% confidence interval (95% CI) of individual NBFRs ranged from 0.43 (95% CI: 0.32, 0.60) to 1.32 (0.92, 1.89) and from 0.34 (0.24, 0.49) to 0.92 (0.56, 1.51) in the Arctic and Antarctic marine food webs, respectively. The TMFs of most congeners were significantly lower than 1, indicating a trophic dilution potential. This is one of the very few investigations on the trophic transfer of NBFRs in remote Arctic and Antarctic marine ecosystems, which provides a basis for exploring the ecological risks of NBFRs in polar regions.

KHCO3-activated high surface area biochar derived from brown algae: A case study for efficient adsorption of Cr(VI) in aqueous solution.

The current study aimed to assess the effectiveness of biochar formed from algae in the removal of Cr(VI) through the process of impregnating brown algae Sargassum hemiphyllum with KHCO3. The synthesis of KHCO3-activated biochar (KBAB-3), demonstrating remarkable adsorption capabilities for Cr(VI), was accomplished utilizing a mixture of brown algae and KHCO3 in a mass ratio of 1:3, followed by calcination at a temperature of 700 °C. Based on the empirical evidence, it can be observed that KBAB-3 shown a significant ability to adsorb Cr(VI) within a range of 60-160 mg g-1 across different environmental conditions. In addition, the KBAB-3 material demonstrated the advantageous characteristic of easy separation, allowing for the continued maintenance of a high efficiency in removing Cr(VI) even after undergoing numerous cycles of reuse. In conclusion, the application of KBAB-3, a novel adsorbent, exhibits considerable prospects for effective removal of Cr(VI) from diverse water sources in the near future.

Boosting acetaminophen degradation in water by peracetic acid activation: A novel approach using chestnut shell-derived biochar at varied pyrolysis temperatures.

In this study, biochar derived from chestnut shells was synthesized through pyrolysis at varying temperatures from 300 °C to 900 °C. The study unveiled that the pyrolysis temperature is pivotal in defining the physical and chemical attributes of biochar, notably its adsorption capabilities and its role in activating peracetic acid (PAA) for the efficient removal of acetaminophen (APAP) from aquatic environments. Notably, the biochar processed at 900 °C, referred to as CN900, demonstrated an exceptional adsorption efficiency of 55.8 mg g-1, significantly outperforming its counterparts produced at lower temperatures (CN300, CN500, and CN700). This enhanced performance of CN900 is attributed to its increased surface area, improved micro-porosity, and a greater abundance of oxygen-containing functional groups, which are a consequence of the elevated pyrolysis temperature. These oxygen-rich functional groups, such as carbonyls, play a crucial role in facilitating the decomposition of the O-O bond in PAA, leading to the generation of reactive oxygen species (ROS) through electron transfer mechanisms. This investigation contributes to the development of sustainable and cost-effective materials for water purification, underscoring the potential of chestnut shell-derived biochar as an efficient adsorbent and catalyst for PAA activation, thereby offering a viable solution for environmental cleanup efforts.

Transcriptome characteristics of epithelial cells from advanced non-small cell lung cancer were revealed by single-cell RNA sequencing.

BACKGROUND: Lung adenocarcinoma (LUAD) is the prevalent malignancy worldwide. The aim is to explore differentially expressed genes (DEGs) associated with immune infiltration and survival time of LUAD patients, and predict transcriptional factors for shedding new light on molecular mechanisms and individual therapy of LUAD. METHOD: ScRNA-seq data of LUAD patients was downloaded from GSE148071 and analyzed by R packages. The clustering and protein-protein interaction network were constructed for screening DEGs. Gene Set Enrichment Analysis (GSEA) and GO enrichment analysis were performed in epithelial cell subgroups with high differentiation potential. Potential regulatory transcription factors were predicted. RESULTS: Sixteen epithelial cell types were required and top 20 genes were identified on cell subgroup Epi4 with the highest differentiation potential associated with poor prognosis of LUAD in PPI network. GSEA and GO annotation results showed that cell subgroup Epi4 was enriched in the biological processes of cell proliferation and energy metabolism, and positively regulated the function of cell proliferation. TPI1 was significantly highly expressed in LUAD samples (p < .0001). TPI1 demonstrated a negative correlation with the infiltration levels of CD8+ T cells, CD4+ T cells, B cells, and activated mast cells, whilst manifesting a positive correlation with the infiltration levels of resident mast cells, Th2 cells, and MDSC. Epi4 was regulated by transcription factors MXD3 and GATA4. CONCLUSION: Overexpression of TPI1 was identified as a novel biomarker for LUAD, and potential regulatory transcription factors MXD3 and GATA4 regulated the proliferation of LUAD with the poor prognosis, which may serve as potential targets to suppress the proliferation of LUAD.

Graphene-based functional electrochemical sensors for the detection of chlorpyrifos in water and food samples: a review.

Prolonged and excessive use of chlorpyrifos (CPS) has caused severe pollution, particularly in crops, vegetables, fruits, and water sources. As a result, CPS is detected in various food and water samples using conventional methods. However, its applications are limited due to size, portability, cost, etc. In this regard, electrochemical sensors are preferred for CPS detection due to their high sensitivity, reliability, rapid, on-site detection, and user-friendly. Notably, graphene-based electrochemical sensors have gained more attention due to their unique physiochemical and electrochemical properties. It shows high sensitivity, selectivity, and quick response because of its high surface area and high conductivity. In this review, we have discussed an overview of three graphene-based different functional electrochemical sensors such as electroanalytical sensors, bio-electrochemical sensors, and photoelectrochemical sensors used to detect CPS in food and water samples. Furthermore, the fabrication and operation of these electrochemical sensors using various materials (low band gap material, nanomaterials, enzymes, antibodies, DNA, aptamers, and so on) and electrochemical techniques (CV, DPV, EIS, SWV etc.) are discussed. The study found that the electrical signal was reduced with increasing CPS concentration. This is due to the blocking of active sites, reduced redox reaction, impedance, irreversible reactions, etc. In addition, acetylcholinesterase-coupled sensors are more sensitive and stable than others. Also, it can be further improved by fabricating with low band gap nanomaterials. Despite their advantages, these sensors have significant drawbacks, such as low reusability, repeatability, stability, and high cost. Therefore, further research is required to overcome such limitations.

Development in health-promoting essential polyunsaturated fatty acids production by microalgae: a review.

Polyunsaturated fatty acids (PUFAs) found in microalgae, primarily omega-3 (ω-3) and omega-6 (ω-6) are essential nutrients with positive effects on diseases such as hyperlipidemia, atherosclerosis, and coronary risk. Researchers still seek improvement in PUFA yield at a large scale for better commercial prospects. This review summarizes advancements in microalgae PUFA research for their cost-effective production and potential applications. Moreover, it discusses the most promising cultivation modes using organic and inorganic sources. It also discusses biomass hydrolysates to increase PUFA production as an alternative and sustainable organic source. For cost-effective PUFA production, heterotrophic, mixotrophic, and photoheterotrophic cultivation modes are assessed with traditional photoautotrophic production modes. Also, mixotrophic cultivation has fascinating sustainable attributes over other trophic modes. Furthermore, it provides insight into growth phase (stage I) improvement strategies to accumulate biomass and the complementing effects of other stress-inducing strategies during the production phase (stage II) on PUFA enhancement under these cultivation modes. The role of an excessive or limiting range of salinity, nutrients, carbon source, and light intensity were the most effective parameter in stage II for accumulating higher PUFAs such as ω-3 and ω-6. This article outlines the commercial potential of microalgae for omega PUFA production. They reduce the risk of diabetes, cardiovascular diseases (CVDs), cancer, and hypertension and play an important role in their emerging role in healthy lifestyle management.

Novel food isolates with striking α-glucosidase inhibitory activity and probiotic potential for an antidiabetic role.

In the current study, ten lactic acid bacteria (LAB) isolates exhibiting anti-α-glucosidase activity were isolated from fermented food. It is directed at novel supplementary diets to prevent/improve diet-induced carbohydrate metabolism disorders and related chronic diseases. Moreover, to evaluate their safety, functionality, and probiotic potential via in vitro simulated test conditions. From 16s-rRNA sequencing, Pediococcus acidilactici (NKUST 803, 845, 858), Lactobacillus plantarum (NKUST 817, 828, 851), Levilactobacillus brevis (NKUST 816, 855) and Lactobacillus acidophilus (NKUST 803, 863) were identified. The results showed that the isolates possessed anti-pathogenic activity, auto-aggregation ability, hydrophobicity (47.44-96.4%), and gastric acid-resistant activity (79-99.1%), which proved their potential for probiotics in nutraceuticals to render hypoglycemic activity or antidiabetic effects to the host positively. Among tested isolates, L. plantarum 817 and P. acidilactici 858 exhibited maximum α-glucosidase inhibitory (AGI) activity of 35-40%. The heat map clearly showed that L. plantarum 817 exhibited the best AGI activity and probiotic potential, among others. These were studied under various simulated gut conditions and safety tests. However, all isolates possess the potential to be used as probiotics in commercial-scale health applications. Pediococcus sp. possesses notable AGI activity but relatively less colonization potential in the gut hence recommended daily intake for positive health effects.

Dynamic molecular switches with hysteretic negative differential conductance emulating synaptic behaviour.

To realize molecular-scale electrical operations beyond the von Neumann bottleneck, new types of multifunctional switches are needed that mimic self-learning or neuromorphic computing by dynamically toggling between multiple operations that depend on their past. Here, we report a molecule that switches from high to low conductance states with massive negative memristive behaviour that depends on the drive speed and number of past switching events, with all the measurements fully modelled using atomistic and analytical models. This dynamic molecular switch emulates synaptic behavior and Pavlovian learning, all within a 2.4-nm-thick layer that is three orders of magnitude thinner than a neuronal synapse. The dynamic molecular switch provides all the fundamental logic gates necessary for deep learning because of its time-domain and voltage-dependent plasticity. The synapse-mimicking multifunctional dynamic molecular switch represents an adaptable molecular-scale hardware operable in solid-state devices, and opens a pathway to simplify dynamic complex electrical operations encoded within a single ultracompact component.

Molecular basis for ubiquitin ligase CRL2FEM1C-mediated recognition of C-degron.

Proteome integrity depends on the ubiquitin-proteasome system to degrade unwanted or abnormal proteins. In addition to the N-degrons, C-terminal residues of proteins can also serve as degradation signals (C-degrons) that are recognized by specific cullin-RING ubiquitin ligases (CRLs) for proteasomal degradation. FEM1C is a CRL2 substrate receptor that targets the C-terminal arginine degron (Arg/C-degron), but the molecular mechanism of substrate recognition remains largely elusive. Here, we present crystal structures of FEM1C in complex with Arg/C-degron and show that FEM1C utilizes a semi-open binding pocket to capture the C-terminal arginine and that the extreme C-terminal arginine is the major structural determinant in recognition by FEM1C. Together with biochemical and mutagenesis studies, we provide a framework for understanding molecular recognition of the Arg/C-degron by the FEM family of proteins.

CT radiomics nomogram for prediction of the Ki-67 index in head and neck squamous cell carcinoma.

OBJECTIVES: To construct and validate a contrast-enhanced computed tomography (CECT)-based radiomics nomogram to predict Ki-67 expression level in head and neck squamous cell carcinoma (HNSCC). METHODS: A total of 217 patients with HNSCC who underwent CECT scans and immunohistochemical examination of their Ki-67 index were enrolled in this study. The patients were divided into a training set (n = 140; Ki-67: ≥ 50% [n = 72] and < 50% [n = 68]) and an external test set (n = 77; Ki-67: ≥ 50% [n = 38] and < 50% [n = 39]). The least absolute shrinkage and selection operator method was used to select key features for a CECT-image-based radiomics signature and a radiomics score (Rad-score) was calculated. A clinical model was established using clinical data and CT findings. The independent clinical factors and Rad-score were then combined to construct a radiomics nomogram. The performance characteristics of the Rad-score, clinical model, and nomogram were assessed using ROCs and decision curve analysis. RESULTS: Twenty features were finally selected to construct the Rad-score. The radiomics nomogram incorporating the Rad-score, low histological grade, and lymphatic spread showed higher predictive value for the Ki-67 index (≥ 50% vs. < 50%) than the clinical model on both the training (AUC, 0.919 vs. 0.648, p < 0.001) and test (AUC, 0.832 vs. 0.685, p = 0.030) sets. Decision curve analysis demonstrated that the radiomics nomogram was more clinically useful than the clinical model. CONCLUSIONS: A CECT-based radiomics nomogram was constructed to predict the expression of Ki-67 in HNSCC. This model showed favorable predictive efficacy and might be useful for prognostic evaluation and clinical decision-making in patients with HNSCC. KEY POINTS: • Accurate pre-treatment prediction of Ki-67 index in HNSCC is crucial. • A CECT-based radiomics nomogram showed favorable predictive efficacy in estimation of Ki-67 expression status in HNSCC patients.

Primary bone tumor detection and classification in full-field bone radiographs via YOLO deep learning model.

OBJECTIVES: Automatic bone lesions detection and classifications present a critical challenge and are essential to support radiologists in making an accurate diagnosis of bone lesions. In this paper, we aimed to develop a novel deep learning model called You Only Look Once (YOLO) to handle detecting and classifying bone lesions on full-field radiographs with limited manual intervention. METHODS: In this retrospective study, we used 1085 bone tumor radiographs and 345 normal bone radiographs from two centers between January 2009 and December 2020 to train and test our YOLO deep learning (DL) model. The trained model detected bone lesions and then classified these radiographs into normal, benign, intermediate, or malignant types. The intersection over union (IoU) was used to assess the model's performance in the detection task. Confusion matrices and Cohen's kappa scores were used for evaluating classification performance. Two radiologists compared diagnostic performance with the trained model using the external validation set. RESULTS: In the detection task, the model achieved accuracies of 86.36% and 85.37% in the internal and external validation sets, respectively. In the DL model, radiologist 1 and radiologist 2 achieved Cohen's kappa scores of 0.8187, 0.7927, and 0.9077 for four-way classification in the external validation set, respectively. The YOLO DL model illustrated a significantly higher accuracy for intermediate bone tumor classification than radiologist 1 (95.73% vs 88.08%, p = 0.004). CONCLUSIONS: The developed YOLO DL model could be used to assist radiologists at all stages of bone lesion detection and classification in full-field bone radiographs. KEY POINTS: • YOLO DL model can automatically detect bone neoplasms from full-field radiographs in one shot and then simultaneously classify radiographs into normal, benign, intermediate, or malignant. • The dataset used in this retrospective study includes normal bone radiographs. • YOLO can detect even some challenging cases with small volumes.