Search

Elevators are essential tools in daily life; timely and accurate fault diagnosis plays a crucial role in ensuring their safe operation. However, the existing elevator fault diagnosis methods often neglect the imbalance between the actual collected normal samples and the fault samples, resulting in low diagnostic accuracy. In this study, we propose an improved Aquila optimizer (IAO) extreme gradient boosting tree (XGBoost)-based elevator fault diagnosis method under unbalanced samples. The proposed method includes three main components: multi-domain feature extraction, sample balancing, and fault diagnosis. In the feature extraction phase, the time domain, frequency domain and entropy features of the vibration signal are extracted. In the sample balance phase, aiming at the problem of unbalanced fault samples, after feature selection using recursive feature elimination (RFE), the minority class samples are oversampled by applying SMOTE-Tomek. In the fault diagnosis phase, IAO is used to optimize the hyperparameters in the XGBoost, and the optimized hyperparameters are brought into XGBoost for fault diagnosis. The fault diagnosis accuracy of the method proposed in this study can reach 99.06%, and the method can accurately identify the fault state of the elevator.

According to the World Health Organization (WHO), gastric cancer (GC) is the fourth leading cause of tumor-related mortality globally and one of the most prevalent malignant tumors. To better understand the role of tumor-infiltrating B cells (TIBs) in GC, this work used single-cell RNA sequencing (scRNA-Seq) and bulk RNA sequencing (bulk RNA-Seq) data to identify candidate hub genes. Both scRNA-Seq and bulk RNA-Seq data for stomach adenocarcinoma (STAD) were obtained from the GEO and TCGA databases, respectively. Using scRNA-seq data, the FindNeighbors and FindClusters tools were used to group the cells into distinct groups. Immune cell clusters were sought in the massive RNA-seq expression matrix using the single-sample gene set enrichment analysis (ssGSEA). The expression profiles were used in Weighted Gene Coexpression Network Analysis (WGCNA) to build TCGA’s gene coexpression networks. Next, univariate Cox regression, LASSO regression, and Kaplan–Meier analyses were used to identify hub genes in scRNA-seq data from sequential B-cell analyses. Finally, we examined the correlation between the hub genes and TIBs utilizing the TISIDB database. We confirmed the immune-related markers in clinical validation samples using reverse transcriptase polymerase chain reaction (RT-PCR) and immunohistochemistry (IHC). 15 cell clusters were classified in the scRNA-seq database. According to the WGCNA findings, the green module is most associated with cancer and B cells. The intersection of 12 genes in two separate datasets (scRNA and bulk) was attained for further analysis. However, survival studies revealed that increased C-X-C motif chemokine receptor 4 (CXCR4) expression was linked to worse overall survival. CXCR4 expression is correlated with active, immature, and memory B cells in STAD were identified. Finally, RT-PCR and IHC assays verified that in GC, CXCR4 is overexpressed, and its expression level correlates with TIBs. We used scRNA-Seq and bulk RNA-Seq to study STAD’s cellular composition. We found that CXCR4 is highly expressed by TIBs in GC, suggesting that it may serve as a hub gene for these cells and a starting point for future research into the molecular mechanisms by which these immune cells gain access to tumors and potentially identify therapeutic targets.

Conotoxins are widely distributed and important for studying ligand-gated ion channels. TxIB, a conotoxin consisting of 16 amino acids derived from Conus textile, is a unique selective ligand that blocks rat α6/α3β2β3 nAChR (IC50 = 28 nM) without affecting other rat subtypes. However, when the activity of TxIB against human nAChRs was examined, it was unexpectedly found that TxIB had a significant blocking effect on not only human α6/α3β2β3 nAChR but also human α6/α3β4 nAChR, with an IC50 of 537 nM. To investigate the molecular mechanism of this species specificity and to establish a theoretical basis for drug development studies of TxIB and its analogs, different amino acid residues between human and rat α6/α3 and β4 nAChR subunits were identified. Each residue of the human species was then substituted with the corresponding residue of the rat species via PCR-directed mutagenesis. The potencies of TxIB towards the native α6/α3β4 nAChRs and their mutants were evaluated through electrophysiological experiments. The results showed that the IC50 of TxIB against h[α6V32L, K61R/α3]β4L107V, V115I was 22.5 μM, a 42-fold decrease in potency compared to the native hα6/α3β4 nAChR. Val-32 and Lys-61 in the human α6/α3 subunit and Leu-107 and Val-115 in the human β4 subunit, together, were found to determine the species differences in the α6/α3β4 nAChR. These results also demonstrate that the effects of species differences between humans and rats should be fully considered when evaluating the efficacy of drug candidates targeting nAChRs in rodent models.

Disease recurrence and metastasis lead to poor prognosis in patients with advanced endometrial carcinoma (EC). RNA-binding proteins (RBPs) are closely associated with tumor initiation and metastasis, but the function and molecular mechanisms of RBPs in EC are unclear. RBPs were screened and identified using the TCGA, GEO, and RBPTD databases. The effect of MEX3A on EC was verified by in vitro and in vivo experiments. Gene set enrichment analysis (GSEA), immunofluorescence (IF), and co-immunoprecipitation (Co-IP) were used to identify potential molecular mechanisms of action. We identified 148 differentially expressed RBPs in EC. MEX3A was upregulated and related to poor prognosis in patients with EC. In vitro and vivo experiments demonstrated that MEX3A promoted the growth, migration, and invasion capacities of EC cells. Mechanistically, DVL3, a positive regulator of the Wnt/β-catenin pathway, also increased the proliferation and metastasis of EC cells. MEX3A enhanced EMT and played a pro-carcinogenic role by interacting with DVL3 to stabilize β-catenin and upregulated the expression of its downstream target genes. MEX3A is upregulated in EC and promotes tumor progression by activating EMT and regulating the Wnt/β-catenin pathway via DVL3. MEX3A may therefore be a novel therapeutic target for EC.

Domain diversities, including inconsistent annotation and varied image collection conditions, inevitably exist among different facial expression recognition (FER) datasets, posing an evident challenge for adapting FER models trained on one dataset to another one. Recent works mainly focus on domain-invariant deep feature learning with adversarial learning mechanisms, ignoring the sibling facial action unit (AU) detection task, which has obtained great progress. Considering that AUs objectively determine facial expressions, this paper proposes an AU-guided unsupervised domain-adaptive FER (AdaFER) framework to relieve the annotation bias between different FER datasets. In AdaFER, we first leverage an advanced model for AU detection on both a source and a target domain. Then, we compare the AU results to perform AU-guided annotating, i.e., target faces that own the same AUs as source faces would inherit the labels from the source domain. Meanwhile, to achieve domain-invariant compact features, we utilize an AU-guided triplet training, which randomly collects anchor–positive–negative triplets on both domains with AUs. We conduct extensive experiments on several popular benchmarks and show that AdaFER achieves state-of-the-art results on all these benchmarks.

In this work, two tailor-made luminogens comprising of electron donors (acridine and phenoxazine) and acceptor (triazine) bridged by the through-space conjugated hexaphenylbenzene (HPB) are synthesized and characterized. Their thermal stability, electrochemical behaviors, crystal, and electronic structures, and photophysical properties are systematically investigated. The crystal and electronic structures reveal that the peripheral phenyls in HPB are closely aligned in a propeller-like fashion, rendering efficient through-space charge transfer between donor and electron moieties. These molecules display weak fluorescence with negligible delayed component in solutions but strong fluorescence with greatly increased delayed component upon aggregate formation, namely aggregation-induced delayed fluorescence (AIDF). Their neat films exhibit high photoluminescence quantum yields (PLQY), and prominent delayed fluorescence. The non-doped organic light-emitting diodes (OLEDs) based on these new luminogens exhibit excellent performance with maximum external quantum efficiency of 12.7% and very small efficiency roll-off of 2.7% at 1,000 cd m−2. Designing AIDF molecules with through-space charge transfer could be a promising strategy to explore robust luminescent materials for efficient non-doped OLEDs.