Computer-aided diagnostic system with automated deep learning method based on the AutoGluon framework improved the diagnostic accuracy of early esophageal cancer.

Background: There have been studies on the application of computer-aided diagnosis (CAD) in the endoscopic diagnosis of early esophageal cancer (EEC), but there is still a significant gap from clinical application. We developed an endoscopic CAD system for EEC based on the AutoGluon framework, aiming to explore the feasibility of automatic deep learning (DL) in clinical application. Methods: The endoscopic pictures of normal esophagus, esophagitis, and EEC were collected from The First Affiliated Hospital of Soochow University (September 2015 to December 2021) and the Norwegian HyperKvasir database. All images of non-cancerous esophageal lesions and EEC in this study were pathologically examined. There were three tasks: task A was normal vs. lesion classification under non-magnifying endoscopy (n=932 vs. 1,092); task B was non-cancer lesion vs. EEC classification under non-magnifying endoscopy (n=594 vs. 429); and task C was non-cancer lesion vs. EEC classification under magnifying endoscopy (n=505 vs. 824). In all classification tasks, we took 100 pictures as the verification set, and the rest comprised as the training set. The CAD system was established based on the AutoGluon framework. Diagnostic performance of the model was compared with that of endoscopists grouped according to years of experience (senior >15 years; junior <5 years). Model evaluation indicators included accuracy, recall rate, precision, F1 value, interpretation time, and the area under the receiver operating characteristic (ROC) curve (AUC). Results: In tasks A and B, the accuracies of medium-performance CAD and high-performance CAD were lower than those of junior doctors and senior doctors. In task C, the medium-performance and high-performance CAD accuracies were close to those of junior doctors and senior doctors. The high-performance CAD model outperformed the junior doctors in both task A (0.850 vs. 0.830) and task C (0.840 vs. 0.830) in sensitivity comparison, but there was still a large gap between high-performance CAD models and doctors in sensitivity comparison. In task A, with the aid of CAD pre-interpretation, the accuracy of junior and senior physicians were significantly improved (from 0.880 to 0.915 and from 0.920 to 0.945, respectively); the time spent on film reading was significantly shortened (junior: from 11.3 to 8.7 s; senior: from 6.7 to 5.5 s). In task C, with the aid of CAD pre-interpretation, the accuracy of junior and senior physicians were significantly improved (from 0.850 to 0.865 and from 0.915 to 0.935, respectively); the reading time was significantly shortened (junior: from 9.5 to 7.7 s; senior: from 5.6 to 3.0 s). Conclusions: The CAD system based on the AutoGluon framework can assist doctors to improve the diagnostic accuracy and reading time of EEC under endoscopy. This study reveals that automatic DL methods are promising in clinical application.

Drymariamides A-J, Antiadipogenic Cyclopeptides Incorporating Noncanonical Amino Acids from Drymaria cordata.

Herein, we report an extensive phytochemical study on the whole plant of Drymaria cordata, which led to the isolation of ten new orbitides, named drymariamides A-J (1-10). Compounds 2, 3, and 5 incorporate rare residues of noncanonical amino acids of kynurenine (Kyn) or 3a-hydroxypyrroloindoline (HPI). Their structures with absolute configurations were elucidated by a combination of spectroscopic analysis, advanced Marfey's method, X-ray diffraction, and electronic circular dichroism analysis. Compounds 1-10 exhibited antiadipogenic effects in 3T3-L1 adipocytes, and the most potent compound 7 showed an EC50 value of 1.17 ± 0.19 µM.

Comparison of brain gray matter volume changes in peritoneal dialysis and hemodialysis patients with chronic kidney disease: a VBM study.

Objective: This study aims to compare gray matter volume changes in patients with chronic kidney disease (CKD) undergoing peritoneal dialysis (PD) and hemodialysis (HD) using voxel-based morphometry (VBM). Methods: A total of 27 PD patients, 25 HD patients, and 42 healthy controls were included. VBM analysis was performed, and cognitive function was assessed using the Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment Scale (MoCA). The correlation between cognitive function and changes in brain gray matter volume was analyzed. Results: Both peritoneal dialysis and hemodialysis patients had partial gray matter volume reduction compared to the controls, but the affected brain regions were not uniform. The hemodialysis patients had greater volume reduction in certain brain regions than the PD patients. The MMSE and MoCA scores were positively correlated with gray matter volume changes. Conclusion: Different dialysis modalities cause damage to specific areas of the brain, which can be detected using VBM. VBM, combined with cognitive function assessment, can help detect structural brain changes and cognitive impairment in patients with different dialysis modalities. The comprehensive application of VBM in the field of neurological function deserves further exploration.

Palladium-Catalyzed Ligand-Controlled Switchable Hetero-(5 + 3)/Enantioselective [2σ+2σ] Cycloadditions of Bicyclobutanes with Vinyl Oxiranes.

For nearly 60 years, significant research efforts have been focused on developing strategies for the cycloaddition of bicyclobutanes (BCBs). However, higher-order cycloaddition and catalytic asymmetric cycloaddition of BCBs have been long-standing formidable challenges. Here, we report Pd-catalyzed ligand-controlled, tunable cycloadditions for the divergent synthesis of bridged bicyclic frameworks. The dppb ligand facilitates the formal (5+3) cycloaddition of BCBs and vinyl oxiranes, yielding valuable eight-membered ethers with bridged bicyclic scaffolds in 100% regioselectivity. The Cy-DPEphos ligand promotes selective hetero-[2σ+2σ] cycloadditions to access pharmacologically important 2-oxabicyclo[3.1.1]heptane (O-BCHeps). Furthermore, the corresponding catalytic asymmetric synthesis of O-BCHeps with 94-99% ee has been achieved using chiral (S)-DTBM-Segphos, representing the first catalytic asymmetric cross-dimerization of two strained rings. The obtained O-BCHeps are promising bioisosteres for ortho-substituted benzenes.

RPI-GGCN: Prediction of RNA-Protein Interaction Based on Interpretability Gated Graph Convolution Neural Network and Co-Regularized Variational Autoencoders.

RNA-protein interactions (RPIs) play an important role in several fundamental cellular physiological processes, including cell motility, chromosome replication, transcription and translation, and signaling. Predicting RPI can guide the exploration of cellular biological functions, intervening in diseases, and designing drugs. Given this, this study proposes the RPI-gated graph convolutional network (RPI-GGCN) method for predicting RPI based on the gated graph convolutional neural network (GGCN) and co-regularized variational autoencoder (Co-VAE). First, different types of feature information were extracted from RNA and protein sequences by nine feature extraction methods. Second, Co-VAEs are used to eliminate the redundancy of fused features and generate optimal features. Finally, this study introduces gated cyclic units into graph convolutional networks (GCNs) to construct a model for RPI prediction, which efficiently extracts topological information and improves the model's interpretable feature learning and expression capabilities. In the fivefold cross-validation test, the RPI-GGCN method achieved prediction accuracies of 97.27%, 97.32%, 96.54%, 95.76%, and 94.98% on the RPI369, RPI488, RPI1446, RPI1807, and RPI2241 datasets. To test the generalization performance of the model, we used the model trained on RPI369 to predict the independent NPInter v3.0 dataset and achieved excellent performance in all six independent validation sets. By visualizing the RPI network graph based on the prediction results, we aim to provide a new perspective and reference for studying RPI mechanisms and exploring new RPIs. Extensive experimental results demonstrate that RPI-GGCN can provide an efficient, accurate, and stable RPI prediction method.

Waveguide-integrated twisted bilayer graphene photodetectors.

Graphene photodetectors have exhibited high bandwidth and capability of being integrated with silicon photonics (SiPh), holding promise for future optical communication devices. However, they usually suffer from a low photoresponsivity due to weak optical absorption. In this work, we have implemented SiPh-integrated twisted bilayer graphene (tBLG) detectors and reported a responsivity of 0.65 A W-1 for telecom wavelength 1,550 nm. The high responsivity enables a 3-dB bandwidth of >65 GHz and a high data stream rate of 50 Gbit s-1. Such high responsivity is attributed to the enhanced optical absorption, which is facilitated by van Hove singularities in the band structure of high-mobility tBLG with 4.1o twist angle. The uniform performance of the fabricated photodetector arrays demonstrates a fascinating prospect of large-area tBLG as a material candidate for heterogeneous integration with SiPh.

Optimization of hydrolysis conditions of alginate based on high performance liquid chromatography.

Alginate is the most abundant polysaccharide compound in brown algae, which is widely used in various fields. At present, the determination of the content of alginate is mostly carried out using sulfuric acid and trifluoroacetic acid hydrolysis followed by the determination of the content, but the results are not satisfactory, and there are problems such as low hydrolysis degree and low recovery rate. Therefore, in this study, based on the optimization of high performance liquid chromatographic conditions for pre-column derivatization of 1-phenyl-3-methyl-5-pyrazolone (PMP), the hydrolysis effects of sulfuric acid, trifluoroacetic acid (TFA), oxalic acid, and formic acid were compared and the hydrolysis conditions were optimized. The results showed that formic acid was the best hydrolyzing acid. The optimal hydrolysis conditions were 95 % formic acid at 110 °C for 10 h. The hydrolysis effect was stable, with high recovery and low destruction of monosaccharides, which made it possible to introduce formic acid into the subsequent polysaccharide hydrolysis. The pre-column derivatization high performance liquid chromatography method established in this study was accurate and reliable, and the hydrolysis acid with better effect was screened, which provided a theoretical basis for the subsequent determination of alginate content.

Effect of Chinese quince proanthocyanidins on the inhibition of heterocyclic amines and quality of fried chicken meatballs and tofu.

Heterocyclic amines (HCAs) have potential carcinogenic and mutagenic activity and are generated in cooked protein-rich foods. Adding proanthocyanidins (PAs) to these foods before frying is an effective way to reduce HCAs. In this study, polymeric PAs (PPA) and ultrasound-assisted acid-catalyzed/catechin nucleophilic depolymerized PAs (UAPA, a type of oligomeric PA) were prepared from Chinese quince fruits (CQF). Different levels of PPA and UAPA (0.05%, 0.1%, and 0.15%) were added to chicken meatballs and tofu; then these foods were fried, and the content of HCAs in them after frying was investigated. The results showed that PPA and, particularly, UAPA significantly inhibited the formation of HCAs in fried meatballs and tofu, and this inhibition was dose-dependent. The inhibition of HCAs by both PPA and UAPA was stronger in the chicken meatballs than in fried tofu. The level of total HCAs was significantly reduced by 57.84% (from 11.93 to 5.03 ng/g) after treatment of meatballs with 0.15% UAPA, with inhibition rates of 78.94%, 50.37%, and 17.81% for norharman, harman, and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), respectively. Of note, there was a negative correlation between water, lipid, protein, creatine, and glucose content and HCA content in the crust, interior, and whole (crust-plus-interior) measurements of all fried samples. Interestingly, PPA and UAPA were found more effective in inhibiting HCAs in the exterior crust than in the interior of the fried chicken meatballs. These results provide evidence that further studies on the reduction of the formation of harmful HCAs in fried foods by adding CQF PAs could be valuable to the fried food industry. PRACTICAL APPLICATION: Chinese quince proanthocyanidins treatments significantly inhibited the generation of heterocyclic amines (HCAs) in chicken meatballs and tofu when deep-fried. These results suggest that Chinese quince proanthocyanidins can be used as natural food additive for reducing HCAs in fried foods, laying the foundation for using Chinese quince fruit proanthocyanidins for HCA inhibition in the food industry.

AI identifies potent inducers of breast cancer stem cell differentiation based on adversarial learning from gene expression data.

Cancer stem cells (CSCs) are a subpopulation of cancer cells within tumors that exhibit stem-like properties and represent a potentially effective therapeutic target toward long-term remission by means of differentiation induction. By leveraging an artificial intelligence approach solely based on transcriptomics data, this study scored a large library of small molecules based on their predicted ability to induce differentiation in stem-like cells. In particular, a deep neural network model was trained using publicly available single-cell RNA-Seq data obtained from untreated human-induced pluripotent stem cells at various differentiation stages and subsequently utilized to screen drug-induced gene expression profiles from the Library of Integrated Network-based Cellular Signatures (LINCS) database. The challenge of adapting such different data domains was tackled by devising an adversarial learning approach that was able to effectively identify and remove domain-specific bias during the training phase. Experimental validation in MDA-MB-231 and MCF7 cells demonstrated the efficacy of five out of six tested molecules among those scored highest by the model. In particular, the efficacy of triptolide, OTS-167, quinacrine, granisetron and A-443654 offer a potential avenue for targeted therapies against breast CSCs.

HydrogelFinder: A Foundation Model for Efficient Self-Assembling Peptide Discovery Guided by Non-Peptidal Small Molecules.

Self-assembling peptides have numerous applications in medicine, food chemistry, and nanotechnology. However, their discovery has traditionally been serendipitous rather than driven by rational design. Here, HydrogelFinder, a foundation model is developed for the rational design of self-assembling peptides from scratch. This model explores the self-assembly properties by molecular structure, leveraging 1,377 self-assembling non-peptidal small molecules to navigate chemical space and improve structural diversity. Utilizing HydrogelFinder, 111 peptide candidates are generated and synthesized 17 peptides, subsequently experimentally validating the self-assembly and biophysical characteristics of nine peptides ranging from 1-10 amino acids-all achieved within a 19-day workflow. Notably, the two de novo-designed self-assembling peptides demonstrated low cytotoxicity and biocompatibility, as confirmed by live/dead assays. This work highlights the capacity of HydrogelFinder to diversify the design of self-assembling peptides through non-peptidal small molecules, offering a powerful toolkit and paradigm for future peptide discovery endeavors.

FBXO22 is a potential therapeutic target for recurrent chondrosarcoma.

Chondrosarcoma (CHS) is a malignant bone tumor with insensitivity to both radiotherapy and chemotherapy, and a high recurrence rate. However, the latent mechanism of recurrent CHS (Re-CHS) remains elusive. Here, we discovered that FBXO22 was highly expressed in clinical samples of Re-CHS. FBXO22 played a significant role in various cancers. However, the role of FBXO22 in Re-CHS remained unclear. Our research demonstrated that suppressing FBXO22 abated the proliferation and migration of CHS cells and facilitated their apoptosis. In addition, suppressing FBXO22 raised the expression of PD-L1 in Re-CHS. All these findings provide new evidence for using FBXO22 and PD-L1 as combined targets to prevent and treat Re-CHS, which may prove to be a novel strategy for immunotherapy of CHS, especially Re-CHS.

SIRT1 maintains bone homeostasis by regulating osteoblast glycolysis through GOT1.

The silent information regulator T1 (SIRT1) is linked to longevity and is a crucial mediator of osteoblast function. We investigated the direct role of Sirt1 during bone modeling and remodeling stages in vivo using Tamoxifen-inducible osteoblast-specific Sirt1 conditional knockout (cKO) mice. cKO mice exhibited lower trabecular and cortical bone mass in the distal femur. These phenotypes were coupled with lower bone formation and bone resorption. Metabolomics analysis revealed that the metabolites involved in glycolysis were significantly decreased in cKO mice. Further analysis of the quantitative acetylome revealed 11 proteins with upregulated acetylation levels in both the femur and calvaria of cKO mice. Cross-analysis identified four proteins with the same upregulated lysine acetylation site in both the femur and calvaria of cKO mice. A combined analysis of the metabolome and acetylome, as well as immunoprecipitation, gene knockout, and site-mutation experiments, revealed that Sirt1 deletion inhibited glycolysis by directly binding to and increasing the acetylation level of Glutamine oxaloacetic transaminase 1 (GOT1). In conclusion, our study suggested that Sirt1 played a crucial role in regulating osteoblast metabolism to maintain bone homeostasis through its deacetylase activity on GOT1. These findings provided a novel insight into the potential targeting of osteoblast metabolism for the treatment of bone-related diseases.

Surface mobility gradient and emergent facilitation in glassy films.

Confining glassy polymers into films can substantially modify their local and film-averaged properties. We present a lattice model of film geometry with void-mediated facilitation behaviors but free from any elasticity effect. We analyze the spatially varying viscosity to delineate the transport properties of glassy films. The film mobility measurements reported by Yang et al., Science, 2010, 328, 1676 are successfully reproduced. The flow exhibits a crossover from a simple viscous flow to a surface-dominated regime as the temperature decreases. The propagation of a highly mobile front induced by the free surface is visualized in real space. Our approach provides a microscopic treatment of the observed glassy phenomena.

Stochastic process is main factor to affect plateau river fish community assembly.

Plateau river ecosystems are often highly vulnerable and responsive to environmental change. The driving mechanism of fish diversity and community assembly in plateau rivers under changing environments presents a significant complexity to the interdisciplinary study of ecology and environment. This study integrated molecular biological techniques and mathematical models to identify the mechanisms influencing spatial heterogeneity of freshwater fish diversity and driving fish community assembly in plateau rivers. By utilizing environmental-DNA metabarcoding and the null model, this study revealed the impact of the stochastic process on fish diversity variations and community assembly in the Huangshui Plateau River of the Yellow River Basin (YRB) in China. This research identified 30 operational taxonomic units (OTUs), which correspond to 20 different fish species. The findings of this study revealed that the fish α-diversity in the upstream region of Xining is significantly higher than in the middle-lower reach (Shannon index: P = 0.017 and Simpson: P = 0.035). This pattern was not found to be related to any other environmental factors besides altitude (P = 0.023) that we measured. Further, the study indicated that the assembly of fish communities in the Huangshui River primarily depends on stochastic ecological processes. These findings suggested that elevation was not the primary factor impacting the biodiversity patterns of fish in plateau rivers. In plateau rivers, spatial heterogeneity of fish community on elevation is mainly determined by stochastic processes under habitat fragmentation, rather than any other physicochemical environmental factors. The limitations of connectivity in the downstream channel of the river could be taken the mainly responsibility for stochastic processes of fish community in Huangshui River. Incorporating ecological processes in the eDNA approach holds great potential for future monitoring and evaluation of fish biodiversity and community assembly in plateau rivers.

Twin prime editing mediated exon skipping/reinsertion for restored collagen VII expression in recessive dystrophic epidermolysis bullosa.

Gene editing nucleases, base editors, and prime editors are potential locus specific genetic treatment strategies for recessive dystrophic epidermolysis bullosa (RDEB); however, many RDEB COL7A1 mutations are unique, making the development of personalized editing reagents challenging. 270 of the ∼320 COL7A1 EB mutations reside in exons that can be skipped, and antisense oligonucleotides (ASO) and gene editing nucleases have been used to create in-frame deletions. ASOs are transient and nucleases generate deleterious double stranded DNA breaks (DSB) and uncontrolled mixtures of allele products. We developed a twin prime editing (twinPE) strategy using the PEmax and recently evolved PE6 prime editors and dual prime editing guide RNAs flanking COL7A1 exon five. Prime editing-mediated deletion of exon 5 with a homozygous premature stop codon was achieved in RDEB fibroblasts, keratinocytes, and iPSC with minimal DSBs, and collagen type VII (C7) protein was restored. TwinPE can replace the target exon with recombinase attachment sequences, and we exploited this to re-insert a normal copy of exon 5 using the Bxb1 recombinase. These findings demonstrate that twinPE can facilitate locus-specific, predictable, in-frame deletions and sequence replacement with few DSBs as a strategy that may enable a single therapeutic agent to treat multiple RDEB patient cohorts.

Exploring Antiviral Drugs on Monolayer Black Phosphorene: Atomistic Theory and Explainable Machine Learning-Assisted Platform.

Favipiravir (FP) and ebselen (EB) belong to a diverse class of antiviral drugs known for their significant efficacy in treating various viral infections. Utilizing molecular dynamics (MD) simulations, machine learning, and van der Waals density functional theory, we accurately elucidate the binding properties of these antiviral drugs on a phosphorene single-layer. To further investigate these characteristics, this study employs four distinct machine learning models-Random Forest, Gradient Boosting, XGBoost, and CatBoost. The Hamiltonian of antiviral molecules within a monolayer of phosphorene is appropriately trained. The key aspect of utilizing machine learning (ML) in drug design revolves around training models that are efficient and precise in approximating density functional theory (DFT). Furthermore, the study employs SHAP (SHapley Additive exPlanations) to elucidate model predictions, providing insights into the contribution of each feature. To explore the interaction characteristics and thermodynamic properties of the hybrid drug, we employ molecular dynamics and DFT calculations in a vacuum interface. Our findings suggest that this functionalized 2D complex exhibits robust thermostability, indicating its potential as an effective and enabled entity. The observed variations in free energy at different surface charges and temperatures suggest the adsorption potential of FP and EB molecules from the surrounding environment.

Theoretical Perspective on the Design of Surface Frustrated Lewis Pairs for Small-Molecule Activation.

The excellent reactivity of frustrated Lewis pairs (FLP) to activate small molecules has gained increasing attention in recent decades. Though the development of surface FLP (SFLP) is prompting the application of FLP in the chemical industry, the design of SFLP with superior activity, high density, and excellent stability for small-molecule activation is still challenging. Herein, we review the progress of designing SFLP by surface engineering, screening natural SFLP, and the dynamic formation of SFLP from theoretical perspectives. We highlight the breakthrough in fine-tuning the activity, density, and stability of the designed SFLP studied by using computational methods. We also discuss future challenges and directions in designing SFLP with outstanding capabilities for small-molecule activation.

Graphdiyne Enabled Nitrogen Vacancy Formation in Copper Nitride for Efficient Ammonia Synthesis.

The electrocatalytic reduction of nitrate is promising for sustainable ammonia synthesis but suffers from slow reduction kinetics and multiple competing reactions. Here, we report a catalyst featuring copper nitride (Cu3N) anchored on a novel graphdiyne support (termed Cu3N/GDY), which is used for electrocatalytic reduction of nitrate to produce ammonia. The GDY absorbed hydrogen and enabled nitrogen (N) vacancy formation in Cu3N for the fast nitrate reduction reaction (NO3RR). Further, the distinct absorption sites formed by GDY and N vacancy enabled the excellent selectivity and stability of NO3RR. Notably, the Cu3N/GDY catalyst achieved a high ammonia yield (YNH3) up to 35280 µg h-1 mgcat.-1 and a high Faradaic efficiency (FE) of 98.1% using 0.1 M NO3- at -0.9 V versus a reversible hydrogen electrode (RHE). Using electron paramagnetic resonance (EPR) technology and in situ X-ray absorption fine structure (XAFS) spectroscopy measurement, we visualized the N vacancy formation in Cu3N and electrocatalytic NO3RR enabled by GDY. These findings show the promise of GDY in sustainable ammonia synthesis and highlight the efficacy of Cu3N/GDY as a catalyst.

Long-Term Outcome of Photobiomodulation Therapy for Refractory Diabetic Wound After Secretory Carcinoma of the Parotid Gland Surgery: A Case Report.

Objective: To evaluate the efficacy and safety of photobiomodulation therapy (PBMT) in the treatment of diabetic patients with refractory wound. Background: Refractory wound is one of the most challenging clinical complications of diabetes mellitus. Studies have shown that PBMT can promote wound healing in many ways. Methods: We reported a 55-year-old male patient with refractory diabetic wound after secretory carcinoma of the parotid gland surgery responding to 810 nm laser. Results: After PBMT, the refractory diabetic wound healed gradually without adverse events. During follow-up 5-years, the healed wound remained stable and showed no signs of recurrence. Conclusions: PBMT can be potentially considered as a therapeutic method in diabetic patients with refractory diabetic wound.