Structure-preserved dimension reduction using joint triplets sampling for multi-batch integration of single-cell transcriptomic data.

Dimension reduction (DR) plays an important role in single-cell RNA sequencing (scRNA-seq), such as data interpretation, visualization and other downstream analysis. A desired DR method should be applicable to various application scenarios, including identifying cell types, preserving the inherent structure of data and handling with batch effects. However, most of the existing DR methods fail to accommodate these requirements simultaneously, especially removing batch effects. In this paper, we develop a novel structure-preserved dimension reduction (SPDR) method using intra- and inter-batch triplets sampling. The constructed triplets jointly consider each anchor's mutual nearest neighbors from inter-batch, k-nearest neighbors from intra-batch and randomly selected cells from the whole data, which capture higher order structure information and meanwhile account for batch information of the data. Then we minimize a robust loss function for the chosen triplets to obtain a structure-preserved and batch-corrected low-dimensional representation. Comprehensive evaluations show that SPDR outperforms other competing DR methods, such as INSCT, IVIS, Trimap, Scanorama, scVI and UMAP, in removing batch effects, preserving biological variation, facilitating visualization and improving clustering accuracy. Besides, the two-dimensional (2D) embedding of SPDR presents a clear and authentic expression pattern, and can guide researchers to determine how many cell types should be identified. Furthermore, SPDR is robust to complex data characteristics (such as down-sampling, duplicates and outliers) and varying hyperparameter settings. We believe that SPDR will be a valuable tool for characterizing complex cellular heterogeneity.

A joint deep learning model enables simultaneous batch effect correction, denoising, and clustering in single-cell transcriptomics.

Recent developments of single-cell RNA-seq (scRNA-seq) technologies have led to enormous biological discoveries. As the scale of scRNA-seq studies increases, a major challenge in analysis is batch effects, which are inevitable in studies involving human tissues. Most existing methods remove batch effects in a low-dimensional embedding space. Although useful for clustering, batch effects are still present in the gene expression space, leaving downstream gene-level analysis susceptible to batch effects. Recent studies have shown that batch effect correction in the gene expression space is much harder than in the embedding space. Methods such as Seurat 3.0 rely on the mutual nearest neighbor (MNN) approach to remove batch effects in gene expression, but MNN can only analyze two batches at a time, and it becomes computationally infeasible when the number of batches is large. Here, we present CarDEC, a joint deep learning model that simultaneously clusters and denoises scRNA-seq data while correcting batch effects both in the embedding and the gene expression space. Comprehensive evaluations spanning different species and tissues showed that CarDEC outperforms Scanorama, DCA + Combat, scVI, and MNN. With CarDEC denoising, non-highly variable genes offer as much signal for clustering as the highly variable genes (HVGs), suggesting that CarDEC substantially boosted information content in scRNA-seq. We also showed that trajectory analysis using CarDEC's denoised and batch-corrected expression as input revealed marker genes and transcription factors that are otherwise obscured in the presence of batch effects. CarDEC is computationally fast, making it a desirable tool for large-scale scRNA-seq studies.

Revealing the Subzero-Temperature Electrochemical Kinetics Behaviors in Ni-Rich Cathode.

The sluggish kinetics in Ni-rich cathodes at subzero temperatures causes decreased specific capacity and poor rate capability, resulting in slow and unstable charge storage. So far, the driving force of this phenomenon remains a mystery. Herein, with the help of in-situ X-ray diffraction and time of flight secondary ion mass spectrometry techniques, the continuous accumulation of both the cathode electrolyte interphase (CEI) film formation and the incomplete structure evolution during cycling under subzero temperature are proposed. It is presented that excessively uniform and thick CEI film generated at subzero temperatures would block the diffusion of Li+ -ions, resulting in incomplete phase evolution and clear charge potential delay. The incomplete phase evolution throughout the Li+ -ion intercalation/de-intercalation processes would further cause low depth of discharge and poor electrochemical reversibility with low initial Coulombic efficiency, as well. In addition, the formation of the thick and uniform CEI film would also consume Li+ -ions during the charging process. This discovery highlights the effects of the CEI film formation behavior and incomplete phase evolution in restricting electrochemical kinetics under subzero temperatures, which the authors believe would promote the further application of the Ni-rich cathodes.

SpaGCN: Integrating gene expression, spatial location and histology to identify spatial domains and spatially variable genes by graph convolutional network.

Recent advances in spatially resolved transcriptomics (SRT) technologies have enabled comprehensive characterization of gene expression patterns in the context of tissue microenvironment. To elucidate spatial gene expression variation, we present SpaGCN, a graph convolutional network approach that integrates gene expression, spatial location and histology in SRT data analysis. Through graph convolution, SpaGCN aggregates gene expression of each spot from its neighboring spots, which enables the identification of spatial domains with coherent expression and histology. The subsequent domain guided differential expression (DE) analysis then detects genes with enriched expression patterns in the identified domains. Analyzing seven SRT datasets using SpaGCN, we show it can detect genes with much more enriched spatial expression patterns than competing methods. Furthermore, genes detected by SpaGCN are transferrable and can be utilized to study spatial variation of gene expression in other datasets. SpaGCN is computationally fast, platform independent, making it a desirable tool for diverse SRT studies.

ScCAEs: deep clustering of single-cell RNA-seq via convolutional autoencoder embedding and soft K-means.

Clustering and cell type classification are a vital step of analyzing scRNA-seq data to reveal the complexity of the tissue (e.g. the number of cell types and the transcription characteristics of the respective cell type). Recently, deep learning-based single-cell clustering algorithms become popular since they integrate the dimensionality reduction with clustering. But these methods still have unstable clustering effects for the scRNA-seq datasets with high dropouts or noise. In this study, a novel single-cell RNA-seq deep embedding clustering via convolutional autoencoder embedding and soft K-means (scCAEs) is proposed by simultaneously learning the feature representation and clustering. It integrates the deep learning with convolutional autoencoder to characterize scRNA-seq data and proposes a regularized soft K-means algorithm to cluster cell populations in a learned latent space. Next, a novel constraint is introduced to the clustering objective function to iteratively optimize the clustering results, and more importantly, it is theoretically proved that this objective function optimization ensures the convergence. Moreover, it adds the reconstruction loss to the objective function combining the dimensionality reduction with clustering to find a more suitable embedding space for clustering. The proposed method is validated on a variety of datasets, in which the number of clusters in the mentioned datasets ranges from 4 to 46, and the number of cells ranges from 90 to 30 302. The experimental results show that scCAEs is superior to other state-of-the-art methods on the mentioned datasets, and it also keeps the satisfying compatibility and robustness. In addition, for single-cell datasets with the batch effects, scCAEs can ensure the cell separation while removing batch effects.

Propensity score matching enables batch-effect-corrected imputation in single-cell RNA-seq analysis.

Developments of single-cell RNA sequencing (scRNA-seq) technologies have enabled biological discoveries at the single-cell resolution with high throughput. However, large scRNA-seq datasets always suffer from massive technical noises, including batch effects and dropouts, and the dropout is often shown to be batch-dependent. Most existing methods only address one of the problems, and we show that the popularly used methods failed in trading off batch effect correction and dropout imputation. Here, inspired by the idea of causal inference, we propose a novel propensity score matching method for scRNA-seq data (scPSM) by borrowing information and taking the weighted average from similar cells in the deep sequenced batch, which simultaneously removes the batch effect, imputes dropout and denoises data in the entire gene expression space. The proposed method is testified on two simulation datasets and a variety of real scRNA-seq datasets, and the results show that scPSM is superior to other state-of-the-art methods. First, scPSM improves clustering accuracy and mixes cells of the same type, suggesting its ability to keep cell type separation while correcting for batch. Besides, using the scPSM-integrated data as input yields results free of batch effects or dropouts in the differential expression analysis. Moreover, scPSM not only achieves ideal denoising but also preserves real biological structure for downstream gene-based analyses. Furthermore, scPSM is robust to hyperparameters and small datasets with a few cells but enormous genes. Comprehensive evaluations demonstrate that scPSM jointly provides desirable batch effect correction, imputation and denoising for recovering the biologically meaningful expression in scRNA-seq data.

Identification of reference genes for gene expression studies among different developmental stages of murine hearts.

BACKGROUND: Real-time quantitative polymerase chain reaction (RT-qPCR) is a widely-used standard assay for assessing gene expression. RT-qPCR data requires reference genes for normalization to make the results comparable. Therefore, the selected reference gene should be highly stable in its expression throughout the experimental datasets. So far, reports about the optimal set of reference genes in murine left ventricle (LV) across embryonic and postnatal stages are few. The objective of our research was to identify the appropriate reference genes in murine LV among different developmental stages. METHODS: We investigated the gene expression profiles of 21 widely used housekeeping genes in murine LV from 7 different developmental stages (almost throughout the whole period of the mouse lifespan). The stabilities of the potential reference genes were evaluated by five methods: GeNorm, NormFinder, BestKeeper, Delta-Ct and RefFinder. RESULTS: We proposed a set of reliable reference genes for normalization of RT-qPCR experimental data in different conditions. Furthermore, our results showed that 6 genes (18S, Hmbs, Ubc, Psmb4, Tfrc and Actb) are not recommended to be used as reference genes in murine LV development studies. The data also suggested that the Rplp0 gene might serve as an optimal reference gene in gene expression analysis. CONCLUSIONS: Our study investigated the expression stability of the commonly used reference genes in process of LV development and maturation. We proposed a set of optimal reference genes that are suitable for accurate normalization of RT-qPCR data in specific conditions. Our findings may be helpful in future studies for investigating the gene expression patterns and mechanism of mammalian heart development.

Single-cell transcriptomic identified HIF1A as a target for attenuating acute rejection after heart transplantation.

Acute rejection (AR) is an important contributor to graft failure, which remains a leading cause of death after heart transplantation (HTX). The regulation of immune metabolism has become a new hotspot in the development of immunosuppressive drugs. In this study, Increased glucose metabolism of cardiac macrophages was found in patients with AR. To find new therapeutic targets of immune metabolism regulation for AR, CD45+ immune cells extracted from murine isografts, allografts, and untransplanted donor hearts were explored by single-cell RNA sequencing. Total 20 immune cell subtypes were identified among 46,040 cells. The function of immune cells in AR were illustrated simultaneously. Cardiac resident macrophages were substantially replaced by monocytes and proinflammatory macrophages during AR. Monocytes/macrophages in AR allograft were more active in antigen presentation and inflammatory recruitment ability, and glycolysis. Based on transcription factor regulation analysis, we found that the increase of glycolysis in monocytes/macrophages was mainly regulated by HIF1A. Inhibition of HIF1A could alleviate inflammatory cells infiltration in AR. To find out the effect of HIF1A on AR, CD45+ immune cells extracted from allografts after HIF1A inhibitor treatment were explored by single-cell RNA sequencing. HIF1A inhibitor could reduce the antigen presenting ability and pro-inflammatory ability of macrophages, and reduce the infiltration of Cd4+ and Cd8a+ T cells in AR. The expression of Hif1α in AR monocytes/macrophages was regulated by pyruvate kinase 2. Higher expression of HIF1A in macrophages was also detected in human hearts with AR. These indicated HIF1A may serve as a potential target for attenuating AR.

Large-Scale Fabrication of Polymer Microcavities with Adjustable Openings and Surface Roughness Regulated by the Polarity of both Seed Surface and Monomers.

Polymer microcavities with adjustable openings and surface roughness are fabricated on a large scale via single-hole poly(glycidyl methacrylate) (PGMA) swelling seed particles. The size of openings of these microcavities can be adjusted by changing the amount of hydrophilic monomer, and the degree of surface roughness is easily regulated relying on the adjustment of the polarity of monomer. Furthermore, the morphology of PGMA/poly(styrene-methacrylic acid) (PGMA/P(S-MAA)) microparticles from microcavity to erythrocyte shape is controlled by the polarity of seed surface. From transmission electron microscopy images of PGMA/P(S-MAA) microparticles, a fresh polymer particle appears in the cavity. To confirm this phenomenon, thermal annealing process in dioxane/water solution is carried out. Considering the flexibility of polymers, the openings and closing of the prepared microparticles are regulated following the increase in volume ratio of dioxane/water. Ball-in-bowl-shaped PGMA/P(S-MAA) microparticles are further presented, which proves secondary nucleation of monomer in the polymerization stage.

Effect of minimizing amount of template by addition of macromolecular crowding agent on preparation of molecularly imprinted monolith.

One of the main challenges in the preparation of molecularly imprinted polymers (MIPs) is the substantial initial amount of template needed because of the requirement of high load capacities for most applications. A new strategy of macromolecular crowding was suggested to solve this problem by reducing the amount of template in the polymerization recipe. In a ternary porogenic system of polystyrene (PS) (crowding agent), tetrahydrofuran, and toluene, an imprinted monolithic column with high porosity and good permeability was synthesized using a mixture of ellagic acid (template), acrylamide, and ethylene glycol dimethacrylate. The effect of polymerization factors, including monomer-template molar ratio and the molecular weight and concentration of PS, on the imprinting effect of the resulting MIP monoliths was systematically investigated. At a high ratio of monomer-template (120:1), the greatest imprinting factor of 32.4 was obtained on the MIP monolith with the aid of macromolecular crowding agent. The PS-based imprinted monolith had imprinting even at the extremely high ratio of functional monomer to template of 1510:1. Furthermore, an off-line solid-phase extraction based on the ground MIP was conducted, and the purification recovery of ellagic acid from pomegranate-rind extract was up to 80 %. In conclusion, this approach based on macromolecular crowding is simple, and is especially valuable for those applications of MIP preparation for which a rare template is used.

Bovine serum albumin surface imprinted polymer fabricated by surface grafting copolymerization on zinc oxide rods and its application for protein recognition.

A novel bovine serum albumin (BSA) surface imprinted polymer based on ZnO rods was synthesized by surface grafting copolymerization. It exhibited an excellent recognition performance to bovine serum albumin. The adsorption capacity and imprinting factor of bovine serum albumin could reach 89.27 mg/g and 2.35, respectively. Furthermore, the fluorescence property of ZnO was used for tracing the process of protein imprinting and it implied the excellent optical sensing property of this material. More importantly, the hypothesis that the surface charge of carrier could affect the imprinting process was confirmed. That is, ZnO with positive surface charge could not only improve the recognition specificity of binding sites to template proteins (pI < 7), but also deteriorate the bindings between sites and non-template proteins (pI > 7). It was also important that the reusability of ZnO@BSA molecularly imprinted polymers was satisfactory. This implied that the poor mechanical/chemical stability of traditional zinc oxide sensors could be solved by the introduction of surface grafting copolymerization. These results revealed that the ZnO@BSA molecularly imprinted polymers are a promising optical/electrochemical sensor element.

A micro-computed tomography study of the relationship between radicular grooves and root canal morphology in mandibular first premolars.

OBJECTIVES: This study aims to explore the correlation between radicular grooves and root canal types by quantitatively detecting the radicular groove of mandibular first premolars using micro-computed tomography. MATERIALS AND METHODS: A total of 127 mandibular first premolars were scanned by micro-computed tomography, and 52 teeth with radicular grooves were identified. Details of root canal type and groove length, depth, and location were analyzed from three-dimensional images. RESULTS: A total of 40.9 % (52/127) of teeth had radicular grooves. Most of the grooves (69.5 %) were located on the mesial surface of the root. The prevalence of radicular grooves in single canals (17.4 %; 15/86) was lower than that in multiple and complex canals (90.2 %; 37/41); this difference was statistically significant (P < 0.001). The mean length and depth of radicular groove in type V (7.7 ± 2.16 and 0.87 ± 0.39 mm, respectively) and other types of canals (6.91 ± 2.67 and 0.63 ± 0.27 mm, respectively) were significantly longer and deeper than type I canals (6.06 ± 2.12 and 0.43 ± 0.14 mm, respectively). CONCLUSIONS: Multiple and complex canals had a higher incidence of radicular grooves and more complicated root morphology than single and simple canals. Therefore, the anatomy of radicular grooves may influence root canal morphology. CLINICAL RELEVANCE: The existence of a radicular groove is closely related to root anatomy and root canal morphology. Anatomical complexity increases the difficulty of root canal treatment and periodontal therapy; therefore, the current data may provide clinicians with a more thorough understanding of the relationship between radicular grooves and root canal morphology.

Dietary fiber may benefit chondrocyte activity maintenance.

The understanding of the link between the gut-bone axis is growing yearly, but the mechanisms involved are not yet clear. Our study analyzed the role of Sestrin2 (SESN2)pathway in the gut-bone axis. We established an osteoarthritis (OA) model in Sprague-Dawley (SD) rats using the anterior cruciate ligament transection (ACLT) procedure, followed by a dietary intervention with varying levels of dietary fiber content for 8 weeks. By 16S rRNA sequencing of the gut microbiota, we found that high dietary fiber (HDF) intake could significantly increase the Bacillota-dominant gut microbiota. Meanwhile, enzyme linked immunosorbent assay (ELISA) and histological analysis showed that intervention with HDF could reduce the degree of bone and joint lesions and inflammation. We hypothesize that HDF increased the dominant flora of Bacillota, up-regulated the expression of SESN2 in knee joint, and reduced gut permeability, thereby reducing systemic inflammatory response and the degree of bone and joint lesions. Therefore, the present study confirms that changes in gut microbiota induced by increased dietary fiber intake delayed the onset of OA by promoting up-regulation of SESN2 expression at the knee joint to maintain chondrocyte activity and reduce synovial inflammation.

PBX/Knotted 1 homeobox-2 (PKNOX2) is a novel regulator of myocardial fibrosis.

Much effort has been made to uncover the cellular heterogeneities of human hearts by single-nucleus RNA sequencing. However, the cardiac transcriptional regulation networks have not been systematically described because of the limitations in detecting transcription factors. In this study, we optimized a pipeline for isolating nuclei and conducting single-nucleus RNA sequencing targeted to detect a higher number of cell signal genes and an optimal number of transcription factors. With this unbiased protocol, we characterized the cellular composition of healthy human hearts and investigated the transcriptional regulation networks involved in determining the cellular identities and functions of the main cardiac cell subtypes. Particularly in fibroblasts, a novel regulator, PKNOX2, was identified as being associated with physiological fibroblast activation in healthy hearts. To validate the roles of these transcription factors in maintaining homeostasis, we used single-nucleus RNA-sequencing analysis of transplanted failing hearts focusing on fibroblast remodelling. The trajectory analysis suggested that PKNOX2 was abnormally decreased from fibroblast activation to pathological myofibroblast formation. Both gain- and loss-of-function in vitro experiments demonstrated the inhibitory role of PKNOX2 in pathological fibrosis remodelling. Moreover, fibroblast-specific overexpression and knockout of PKNOX2 in a heart failure mouse model induced by transverse aortic constriction surgery significantly improved and aggravated myocardial fibrosis, respectively. In summary, this study established a high-quality pipeline for single-nucleus RNA-sequencing analysis of heart muscle. With this optimized protocol, we described the transcriptional regulation networks of the main cardiac cell subtypes and identified PKNOX2 as a novel regulator in suppressing fibrosis and a potential therapeutic target for future translational studies.

The Fgfr2(S252W/+) mutation in mice retards mandible formation and reduces bone mass as in human Apert syndrome.

Apert syndrome is a common craniosynostosis caused by gain-of-function missense mutations of fibroblast growth factor receptor 2 (FGFR2). Mice with the FGFR2 S252W mutation can elucidate the mechanism by which the human Apert syndrome phenotypes arise. However, many studies have focused on mutant skull and long bone malformation, only few studies have focused on mandible changes. Bone formation and micro-architecture between 28- and 56-day-old mutant mice and controls were compared to investigate the changes in the mandibular micro-architecture caused by the Fgfr2(S252W/+) mutation to provide a basis for exploring the pathogenesis and therapeutic measures of human Apert syndrome. Fgfr2(S252W/+) mutant mice were established, and their general characteristics, including weight, naso-anal length, and calcium and phosphate content in serum and bone were tested. Calcein labeling, tartrate-resistant acid phosphatase staining and toluidine blue staining were used to detect osteoblast and osteoclast activities. H&E staining and micro-CT detection were used to test micro-architecture changes. The changes in mineral apposition rate and micro-architecture of the Fgfr2(S252W/+) mice were statistically significant; however, the magnitude of the micro-architecture became less with age. The Fgfr2(S252W/+) mutation may retard mandibular bone formation, decreased bone volume, and compromised skeletal architecture by regulating both osteoblastogenesis and osteoclastogenesis.

A micro-computed tomography study of the root canal morphology of the mandibular first premolar in a population from southwestern China.

OBJECTIVES: This study aimed to investigate the root canal morphology of mandibular first premolar teeth in a population from southwestern China by micro-computed tomography (micro-CT). MATERIALS AND METHODS: Human mandibular first premolars (115) were selected and prepared for micro-CT analysis with a slice thickness of 30 µm. Details of root canal orifices, canals, accessory canals, apical foramina-apical delta intercanal communication, loops and isthmuses, and mesial invagination were analyzed from reconstructed three-dimensional (3D) images. RESULTS: Canal patterns categorized according to the classification defined by Vertucci (Endod Top 10:3-29, 2005) as types I (65.2%), III (2.6%), V (22.6%), and VII were identified (0.9%). Accessory canals were present in 35.7% of the samples and were predominantly located in the apical third of the root. A single apical foramen was observed in 50.4% of the samples and two or three foramina in 28.7% and 14.8%, respectively. Apical delta was identified in 6.1% of the samples and the prevalence of intercanal communication and loops was 3.5% and 7%, respectively. Mesial invagination of the root was identified in 27.8% of the samples, the majority of which contained multiple canals. CONCLUSIONS: The data obtained in this study revealed complex root morphology with high prevalence of multiple canals, more than half of which exhibited type I canal patterns. CLINICAL RELEVANCE: Micro-CT was used as a noninvasive technique for 3D investigation of root canal morphology in the mandibular first premolars of a population from southwestern China. Furthermore, data obtained revealed complex anatomy of various types.

Batch alignment of single-cell transcriptomics data using deep metric learning.

scRNA-seq has uncovered previously unappreciated levels of heterogeneity. With the increasing scale of scRNA-seq studies, the major challenge is correcting batch effect and accurately detecting the number of cell types, which is inevitable in human studies. The majority of scRNA-seq algorithms have been specifically designed to remove batch effect firstly and then conduct clustering, which may miss some rare cell types. Here we develop scDML, a deep metric learning model to remove batch effect in scRNA-seq data, guided by the initial clusters and the nearest neighbor information intra and inter batches. Comprehensive evaluations spanning different species and tissues demonstrated that scDML can remove batch effect, improve clustering performance, accurately recover true cell types and consistently outperform popular methods such as Seurat 3, scVI, Scanorama, BBKNN, Harmony et al. Most importantly, scDML preserves subtle cell types in raw data and enables discovery of new cell subtypes that are hard to extract by analyzing each batch individually. We also show that scDML is scalable to large datasets with lower peak memory usage, and we believe that scDML offers a valuable tool to study complex cellular heterogeneity.

Use of a novel tapered, scoring balloon to treat juxta-anastomotic stenosis in AV fistula.

BACKGROUND: Juxta-anastomotic stenosis, the most common lesion in arteriovenous fistula, creates a dilemma for optimal balloon size choosing during PTA. OBJECTIVES: To descript the effect of a novel tapered scoring balloon catheter (DK Medtech, Suzhou, China) which has a conical shape with an increasing diameter from the front part of the balloon to the end, and three non-slip elements attached on the surface of the balloon. RESEARCH DESIGN: Case series of 10 patients used this balloon catheter was retrospectively analyzed. SUBJECTS: Patients with juxta-anastomotic stenosis. MEASURES: A retrospective review of 10 cases using the novel tapered, scoring balloon catheter in our center was performed. RESULTS: All cases were clinical technique success. The average total procedure time was 16.7 min with 2:14 of fluoroscopy time. No complications such as vascular rupture or dissection occurred. The primary patency rates at 6 and 12 month were 80% and 50% separately. CONCLUSIONS: This tapered scoring balloon provides an economical, safe, and efficient management for juxta-anastomotic stenosis.

CuCl2 -Catalyzed α-Chloroketonation of Aromatic Alkenes via Visible-Light-Induced LMCT.

Here we report a copper-catalyzed protocol for the synthesis of α-chloroketones from aromatic alkenes including electron-deficient olefins under visible-light irradiation. Preliminary mechanistic studies show that the peroxo Cu(II) species is the key intermediate and hydroperoxyl (HOO⋅) and chlorine (Cl⋅) radicals can be generated by ligand-to-metal charge transfer (LMCT).

Clinical Efficacy of Endoscopic Submucosal Dissection for the Treatment of Duodenal Lesions in Terms of Operative Technique and Management of Complications.

Background and Aims: Duodenal endoscopic submucosal dissection (ESD) has been considered to be the most challenging because of its high incidence of complications, which has hindered the development of duodenal ESD. The aim of this study is to discuss operation tips for duodenal ESD and to assess the efficacy and safety of duodenal ESD. Patients and Methods: Eighty-two patients who underwent ESD in the digestive endoscope center for superficial duodenal epithelial tumors (SDETs) from January 2017 to June 2021 were studied. Patients were divided into three groups according to the occurrence of complications, and the clinical characteristics and surgical efficacy of each group were compared. Results: SDETs in 82 patients were completely removed by ESD, with a 97.5% R0 resection rate. The average size of resected lesions was 23.8 ± 6.5 mm. There were significant differences in lesion size and operation time between the normal and intraprocedural complication groups (P < .05). Similarly, between the normal and delayed complication groups, significant differences were noted in lesion location, size, operation time, occupied circumference, and postoperative hospitalization duration (P < .05). Conclusion: Duodenal ESD is prone to complications that increase the complexity of the procedure. By improving the necessary technique and skills, duodenal ESD remains safe and effective.