RESUMO
A self-reinforcing positive feedback is regarded as a critical process for maintaining alternative stable states (ASS); however, identification of ASS and quantification of positive feedbacks remain elusive in natural ecosystems. Here, we used large-scale field surveys to search for ASS and a positive feedback mechanism under a wide range of habitats on the Tibetan Plateau. Using multiple methods, we proved that three stable states exist that accompany alpine marsh degradation. Positive feedbacks between changing soil moisture and plant community composition forced the ecosystem into another stable state, and the alteration of water use efficiency (WUE) of the component species contributed to this shift. This study provides the first empirical evidence that positive feedback loops maintain ASS in the alpine marsh ecosystem on the Tibetan Plateau. Our research revealed the powerful driving role of plants in transitions between states, which may support the conservation and restoration of global alpine marsh ecosystems.
Assuntos
Solo , Áreas Alagadas , Solo/química , Tibet , Água , Plantas , EcossistemaRESUMO
Epidermal growth factor receptor (EGFR) is a validated target for non-small-cell lung cancer (NSCLC). However, the treatment for EGFR-C797S mutation induced by third-generation EGFR inhibitors remains a concern. Therefore, the development of the fourth-generation EGFR inhibitors to overcome the EGFR-C797S mutation has great potential for clinical treatment. In this article, we designed and synthesized a series of diphenyl ether substituted quinazolin-4-amine derivatives that simultaneously occupy the ATP binding pocket and the allosteric site of EGFR. Among the newly synthesized compounds, 9d displayed excellent kinase activity against EGFRL858R/T790M/C797S with an IC50 value of 0.005 µM, and exhibited anti-proliferation activity in BaF3-EGFRL858R/T790M/C797S cells with the IC50 value of 0.865 µM. Furthermore, 9d could suppress phosphorylation of EGFR and induce cell apoptosis and cycle arrest at G2 phase in a dose-dependent manner in BaF3-EGFRL858R/T790M/C797S cells. More importantly, 9d displayed significant antitumor effects in BaF3-EGFRL858R/T790M/C797S xenograft mouse model (30 mg/kg, TGI = 71.14 %). All the results indicated compound 9d might be a novel fourth-generation EGFR inhibitor for further development in overcoming the EGFR-C797S resistance mutation.
RESUMO
High levels of extracellular adenosine in tumor microenvironment (TME) has extensive immunosuppressive effect. CD73 catalyzes the conversion of AMP into adenosine and regulates its production. Inhibiting CD73 can reduce the level of adenosine and reverse adenosine-mediated immune suppression. Therefore, CD73 has emerged as a valuable target for cancer immunotherapy. Here, a new series of malonic acid non-nucleoside derivatives were designed, synthesized and evaluated as CD73 inhibitors. Among them, compounds 18 and 19 exhibited significant inhibition activities against hCD73 with IC50 values of 0.28 µM and 0.10 µM, respectively, suggesting the feasibility of replacing the benzotriazole moiety in the lead compound. This study explored the novelty and structural diversity of CD73 inhibitors.
Assuntos
5'-Nucleotidase , Desenho de Fármacos , Inibidores Enzimáticos , Malonatos , Relação Estrutura-Atividade , 5'-Nucleotidase/antagonistas & inibidores , 5'-Nucleotidase/metabolismo , Humanos , Malonatos/química , Malonatos/farmacologia , Malonatos/síntese química , Inibidores Enzimáticos/farmacologia , Inibidores Enzimáticos/síntese química , Inibidores Enzimáticos/química , Estrutura Molecular , Relação Dose-Resposta a Droga , Proteínas Ligadas por GPI/antagonistas & inibidores , Proteínas Ligadas por GPI/metabolismoRESUMO
Underwater imaging technology plays a pivotal role in marine exploration and reconnaissance, necessitating photodetectors (PDs) with high responsivity, fast response speed, and low preparation costs. This study presents the synergistic optimization of responsivity and response speed in self-powered photoelectrochemical (PEC)-type photodetector arrays based on oxygen-vacancy-tuned amorphous gallium oxide (a-Ga2O3) thin films, specifically designed for solar-blind underwater detection. Utilizing a low-cost one-step sputtering process with controlled oxygen flow, a-Ga2O3 thin films with varying oxygen vacancy (VO) concentrations are fabricated. By balancing the trade-offs among electrocatalytic reactions, charge transfer, carrier recombination, and trapping, both the responsivity and response speed of a-Ga2O3-based self-powered PEC-PDs are simultaneously improved. Consequently, the optimized PEC-PDs demonstrated exceptional performance, achieving a responsivity of 33.75 mA W-1 and response times of 12.8 ms (rise) and 31.3 ms (decay), outperforming the vast majority of similar devices. Furthermore, a pronounced positive correlation between anomalous transient photocurrent spikes and the concentration of VO defects is observed, offering compelling evidence for VO-mediated indirect recombination. Finally, the proof-of-concept solar-blind underwater imaging system, utilizing an array of self-powered PEC-PDs, demonstrated clear imaging capabilities in seawater. This work provides valuable insight into the potential for developing cost-effective, high-performance a-Ga2O3 thin-film-based PEC-PDs for advanced underwater imaging technology.
RESUMO
In this study, we present an approach to neuropharmacological research by integrating few-shot meta-learning algorithms with brain activity mapping (BAMing) to enhance the discovery of central nervous system (CNS) therapeutics. By utilizing patterns from previously validated CNS drugs, our approach facilitates the rapid identification and prediction of potential drug candidates from limited datasets, thereby accelerating the drug discovery process. The application of few-shot meta-learning algorithms allows us to adeptly navigate the challenges of limited sample sizes prevalent in neuropharmacology. The study reveals that our meta-learning-based convolutional neural network (Meta-CNN) models demonstrate enhanced stability and improved prediction accuracy over traditional machine-learning methods. Moreover, our BAM library proves instrumental in classifying CNS drugs and aiding in pharmaceutical repurposing and repositioning. Overall, this research not only demonstrates the effectiveness in overcoming data limitations but also highlights the significant potential of combining BAM with advanced meta-learning techniques in CNS drug discovery.
RESUMO
BACKGROUND: The traditional Chinese medicine (TCM) Xiaoliu Pingyi recipe (XLPYR) has been clinically used for several decades, demonstrating favorable therapeutic effects. However, the underlying regulatory mechanisms remain unclear. The aim of this study was to explore the anti-tumor effects of XLPYR and its regulatory role in the vascular microenvironment through in vivo and in vitro experiment. MATERIALS AND METHODS: In the in vivo study, a C57BL/6J mouse model of lung adenocarcinoma (LUAD) allografts was established, and various interventions were administered for 14 days (Model group: administered normal saline via oral gavage; Pemetrexed (PEM) group: intraperitoneally injected with a solution of pemetrexed, once every 3d; XLPYR group: administered XLPYR via oral gavage; Combination (COMBI) group: received XLPYR via oral gavage simultaneously with intraperitoneal injection of pemetrexed solution). Tumor volume and weight were then compared among the groups. The impact of XLPYR on the tumor vascular microenvironment was assessed using immunohistochemistry staining. In the in vitro study, XLPYR-containing serum was prepared by oral administration to SD rats. The CCK-8 assay evaluated the effect of the serum on the proliferation of normal lung epithelial BEAS-2B cells and LUAD A549 cells, determining the optimal intervention concentrations. The cell migration and invasion abilities were evaluated using the wound-healing assay and Transwell assay, respectively. Finally, ELISA assay measured VEGF secretion levels in the LUAD cell supernatant, and RT-qPCR and Western Blot were employed to detect differences in HIF-1α, VEGFA, Ang-2, and PI3K/Akt mRNA and protein expression levels in both in vivo and in vitro experiments. RESULTS: In the in vivo study, XLPYR significantly inhibited the growth of mice LUAD allografts, with enhanced anti-tumor effects observed with prolonged drug intervention. Immunohistochemistry staining revealed reduced MVD and increased pericyte coverage in all intervention groups. Regarding vascular function, FITC-Dextran extravasation in the tumor tissues of the Model group was significantly higher than in the intervention groups, particularly with lower extravasation in the COMBI group compared to the PEM group. In the in vitro study, XLPYR demonstrated a time- and concentration-dependent inhibitory effect on LUAD cells, and with greater sensitivity in inhibiting LUAD cells compared to BEAS-2B cells. The wound-healing assay and Transwell assay confirmed that XLPYR significantly suppressed the migration and invasion abilities of LUAD cells. ELISA experiments further revealed a significant decrease in VEGF expression in the supernatant of each intervention group. RT-qPCR and Western Blot results showed consistent findings between the in vivo and in vitro experiments. HIF-1α, VEGFA, and Ang-2 mRNA and protein expression levels were significantly downregulated in the PEM group, XLPYR group, and COMBI group. There were no significant differences in the expression of PI3K and Akt mRNA and total protein, but the expression levels of phosphorylated p-PI3K and p-Akt were notably downregulated. CONCLUSION: XLPYR significantly inhibited C57BL/6J mouse LUAD allograft growth and improved the vascular microenvironment, thereby intervening in tumor angiogenesis and inducing vascular normalization. It suppressed LUAD cell proliferation, migration, and invasion, while reducing VEGF concentration in the cell supernatant. The regulatory mechanism may involve inhibiting PI3K/Akt protein phosphorylation and downregulating angiogenesis-related factors, such as HIF-1α, VEGF, and Ang-2.
Assuntos
Adenocarcinoma de Pulmão , Medicamentos de Ervas Chinesas , Neoplasias Pulmonares , Camundongos Endogâmicos C57BL , Microambiente Tumoral , Animais , Medicamentos de Ervas Chinesas/farmacologia , Microambiente Tumoral/efeitos dos fármacos , Camundongos , Adenocarcinoma de Pulmão/tratamento farmacológico , Neoplasias Pulmonares/tratamento farmacológico , Neoplasias Pulmonares/patologia , Humanos , Linhagem Celular Tumoral , Proliferação de Células/efeitos dos fármacos , Fator A de Crescimento do Endotélio Vascular/metabolismo , Masculino , Pemetrexede/farmacologia , Neovascularização Patológica/tratamento farmacológico , Movimento Celular/efeitos dos fármacos , Medicina Tradicional Chinesa/métodosRESUMO
Targeting the PD-1/PD-L1 axis with small-molecular inhibitors is a promising approach for immunotherapy. Here, we identify a natural pentacyclic triterpenoid, Pygenic Acid A (PA), as a PD-1 signaling inhibitor. PA exerts anti-tumor activity in hPD-1 knock-in C57BL/6 mice and enhances effector functions of T cells to promote immune responses by disrupting the PD-1 signaling transduction. Furthermore, we identify SHP-2 as the direct molecular target of PA for inhibiting the PD-1 signaling transduction. Subsequently, mechanistic studies suggest that PA binds to a new druggable site in the phosphorylated PD-1 ITSM recognition site of SHP-2, inhibiting the recruitment of SHP-2 by PD-1. Taken together, our findings demonstrate that PA has a potential application in cancer immunotherapy and occupying the phosphorylated ITSM recognition site of SHP-2 may serve as an alternative strategy to develop PD-1 signaling inhibitors. In addition, our success in target recognition provides a paradigm of target identification and confirmation for natural products.
RESUMO
During the drug discovery and design process, the acid-base dissociation constant (pKa) of a molecule is critically emphasized due to its crucial role in influencing the ADMET (absorption, distribution, metabolism, excretion, and toxicity) properties and biological activity. However, the experimental determination of pKa values is often laborious and complex. Moreover, existing prediction methods exhibit limitations in both the quantity and quality of the training data, as well as in their capacity to handle the complex structural and physicochemical properties of compounds, consequently impeding accuracy and generalization. Therefore, developing a method that can quickly and accurately predict molecular pKa values will to some extent help the structural modification of molecules, and thus assist the development process of new drugs. In this study, we developed a cutting-edge pKa prediction model named GR-pKa (Graph Retention pKa), leveraging a message-passing neural network and employing a multi-fidelity learning strategy to accurately predict molecular pKa values. The GR-pKa model incorporates five quantum mechanical properties related to molecular thermodynamics and dynamics as key features to characterize molecules. Notably, we originally introduced the novel retention mechanism into the message-passing phase, which significantly improves the model's ability to capture and update molecular information. Our GR-pKa model outperforms several state-of-the-art models in predicting macro-pKa values, achieving impressive results with a low mean absolute error of 0.490 and root mean square error of 0.588, and a high R2 of 0.937 on the SAMPL7 dataset.
Assuntos
Redes Neurais de Computação , Termodinâmica , Descoberta de Drogas/métodosRESUMO
Traditional optical communication systems rely on single narrow-band PDs, which can expose confidential information and data to potential eavesdropping in free space. With advancements in technology, even optical communication in the UV spectrum, invisible to the sun, faces risks of interception. Consequently, broad-band PDs that combine optical encryption with algorithmic encryption hold significant promise for secure and reliable communication. This study presents a photodetector based on TiO2-α-Ga2O3 heterostructures, prepared via direct oxidation and hydrothermal reaction, demonstrating self-powered UVC/UVA broad-band detection capabilities. The PD exhibits response peaks at approximately 250 and 320 nm, with R of 42.16 and 59.88 mA/W and D* of 8.21 × 1013 and 9.56 × 1013 Jones, respectively. Leveraging the superior optical response characteristics of UVC and UVA wavelengths, this device has been employed to develop a communication system designed for data transmission. The proposed system features two independent channels: one for data transmission using UVC and another for key distribution using UVA. Secure communication is ensured through specialized encryption algorithms. In summary, this work offers a straightforward, cost-effective, and practical method for fabricating self-powered UVC/UVA broad-band PDs. This PD provides new insights into the development of multi-purpose, multi-band secure optical communication devices and holds promise for integration into multifunctional optoelectronic systems in the future.
RESUMO
Pulmonary fibrosis (PF) is a chronic, progressive and irreversible interstitial lung disease characterized by unremitting pulmonary myofibroblasts activation, extracellular matrix (ECM) deposition and inflammatory recruitment. PF has no curable medication yet. In this study we investigated the molecular pathogenesis and potential therapeutic targets of PF and discovered drug lead compounds for PF therapy. A murine PF model was established in mice by intratracheal instillation of bleomycin (BLM, 5 mg/kg). We showed that the protein level of pulmonary protein phosphatase magnesium-dependent 1A (PPM1A, also known as PP2Cα) was significantly downregulated in PF patients and BLM-induced PF mice. We demonstrated that TRIM47 promoted ubiquitination and decreased PPM1A protein in PF progression. By screening the lab in-house compound library, we discovered otilonium bromide (OB, clinically used for treating irritable bowel syndrome) as a PPM1A enzymatic activator with an EC50 value of 4.23 µM. Treatment with OB (2.5, 5 mg·kg-1·d-1, i.p., for 20 days) significantly ameliorated PF-like pathology in mice. We constructed PF mice with PPM1A-specific knockdown in the lung tissues, and determined that by targeting PPM1A, OB treatment suppressed ECM deposition through TGF-ß/SMAD3 pathway in fibroblasts, repressed inflammatory responses through NF-κB/NLRP3 pathway in alveolar epithelial cells, and blunted the crosstalk between inflammation in alveolar epithelial cells and ECM deposition in fibroblasts. Together, our results demonstrate that pulmonary PPM1A activation is a promising therapeutic strategy for PF and highlighted the potential of OB in the treatment of the disease.
RESUMO
The accurate diagnosis of non-obstructive azoospermia (NOA) and obstructive azoospermia (OA) is crucial for selecting appropriate clinical treatments. This study aimed to investigate the pivotal role of miRNAs in circulating plasma extracellular vesicles (EVs) in distinguishing between NOA and OA, as well as uncovering the signaling pathways involved in azoospermia pathogenesis. In this study, differential expression of EV miR-513c-5p and miR-202-5p was observed between NOA and OA patients, while the selenocompound metabolism pathway could be affected in azoospermia through Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analysis. The predictive power of these microRNAs was evaluated using ROC-AUC analysis, demonstrating promising sensitivity, specificity, and area under the curve values. A binomial regression equation incorporating circulating plasma levels of EVs miR-202-5p and miR-513c-5p along with follicle-stimulating hormone was calculated to provide a clinically applicable method for diagnosing NOA and OA. This study presents a potentially non-invasive testing approach for distinguishing between NOA and OA, offering a possibly valuable tool for clinical practice.
RESUMO
Accurate image classification and retrieval are of importance for clinical diagnosis and treatment decision-making. The recent contrastive language-image pre-training (CLIP) model has shown remarkable proficiency in understanding natural images. Drawing inspiration from CLIP, pathology-dedicated CLIP (PathCLIP) has been developed, utilizing over 200,000 image and text pairs in training. While the performance the PathCLIP is impressive, its robustness under a wide range of image corruptions remains unknown. Therefore, we conduct an extensive evaluation to analyze the performance of PathCLIP on various corrupted images from the datasets of osteosarcoma and WSSS4LUAD. In our experiments, we introduce eleven corruption types including brightness, contrast, defocus, resolution, saturation, hue, markup, deformation, incompleteness, rotation, and flipping at various settings. Through experiments, we find that PathCLIP surpasses OpenAI-CLIP and the pathology language-image pre-training (PLIP) model in zero-shot classification. It is relatively robust to image corruptions including contrast, saturation, incompleteness, and orientation factors. Among the eleven corruptions, hue, markup, deformation, defocus, and resolution can cause relatively severe performance fluctuation of the PathCLIP. This indicates that ensuring the quality of images is crucial before conducting a clinical test. Additionally, we assess the robustness of PathCLIP in the task of image-to-image retrieval, revealing that PathCLIP performs less effectively than PLIP on osteosarcoma but performs better on WSSS4LUAD under diverse corruptions. Overall, PathCLIP presents impressive zero-shot classification and retrieval performance for pathology images, but appropriate care needs to be taken when using it.
RESUMO
In the synthetic laboratory, researchers typically rely on nuclear magnetic resonance (NMR) spectra to elucidate structures of synthesized products and confirm whether they match the desired target compounds. As chemical synthesis technology evolves toward intelligence and continuity, efficient computer-assisted structure elucidation (CASE) techniques are required to replace time-consuming manual analysis and provide the necessary speed. However, current CASE methods typically aim to derive precise chemical structures from spectroscopic data, yet they suffer from drawbacks such as low accuracy, high computational cost, and reliance on chemical libraries. In meticulously designed chemical synthesis reactions, researchers prioritize confirming the attainment of the target product based on NMR spectra, rather than focusing on identifying the specific product obtained. For this purpose, we innovatively developed a binary classification model, termed as MatCS, to directly predict the relationship between NMR spectra image (including 1H NMR and 13C NMR) and the molecular structure of the target compound. After evaluating various feature extraction methods, MatCS employs a combination of the Graph Attention Networks and Graph Convolutional Networks to learn the structural features of molecular graphs and the pretrained ResNet101 network with a Convolutional Block Attention Module to extract features from NMR spectra images. The results show that on a challenging Testsim data set, which poses difficulty in distinguishing spectra of similar molecular structures, MatCS achieves comprehensive evaluation metrics with an F1-score of 0.81 and an AUC value of 0.87. Simultaneously, it exhibited commendable performance on an external SDBS data set containing experimental NMR spectra, showcasing substantial potential for structural verification tasks in real automated chemical synthesis.
Assuntos
Aprendizado Profundo , Espectroscopia de Ressonância Magnética , Espectroscopia de Ressonância Magnética/métodosRESUMO
Objective: The objective of this study was to investigate the impact of transforming growth factor ß1 (TGF-ß1) on epithelial development using an ex vivo model of submandibular gland (SMG) epithelial-mesenchymal separation. Materials and methods: The ex vivo model was established by separating E13 mouse SMG epithelia and mesenchyme, culturing them independently for 24 h, recombining them, and observing branching morphogenesis. Microarray analysis was performed to evaluate the transcriptome of epithelia treated with and without 1 ng/ml TGF-ß1. Differential gene expression, pathway enrichment, and protein-protein interaction networks were analyzed. Quantitative real-time polymerase chain reaction, Western blot, and immunofluorescence were employed to validate the mRNA and protein levels. Results: Recombined SMGs using separated epithelia and mesenchyme that were cultured for 24 h showed a significant inhibition of epithelial development compared to SMGs recombined immediately after separation. The level of TGF-ß1 decreased in the SMG epithelia after epithelia-mesenchyme separation. Epithelia that were separated from mesenchyme for 24 h and pretreated with 1 ng/ml TGF-ß1 continued to develop after recombination with mesenchyme, while epithelia without 1 ng/ml TGF-ß1 treatment did not. Microarray analysis suggested pathway enrichment related to epithelial development and an upregulation of Sox2 in the 1 ng/ml TGF-ß1-treated epithelia. Further experiments validated the phosphorylation of SMAD2 and SMAD3, upregulation of SOX2 and genes associated with epithelial development, including Prol1, Dcpp1, Bhlha15, Smgc, and Bpifa2. Additionally, 1 ng/ml TGF-ß1 inhibited epithelial apoptosis by improving the BCL2/BAX ratio and reducing cleaved caspase 3. Conclusions: The addition of 1 ng/ml TGF-ß1 maintained the developmental potential of embryonic SMG epithelia separated from mesenchyme for 24 h. This suggests that 1 ng/ml TGF-ß1 may partially compensate for the role of mesenchyme during the separation phase, although its compensation is limited in extent.
RESUMO
Analyzing drug-related interactions in the field of biomedicine has been a critical aspect of drug discovery and development. While various artificial intelligence (AI)-based tools have been proposed to analyze drug biomedical associations (DBAs), their feature encoding did not adequately account for crucial biomedical functions and semantic concepts, thereby still hindering their progress. Since the advent of ChatGPT by OpenAI in 2022, large language models (LLMs) have demonstrated rapid growth and significant success across various applications. Herein, LEDAP was introduced, which uniquely leveraged LLM-based biotext feature encoding for predicting drug-disease associations, drug-drug interactions, and drug-side effect associations. Benefiting from the large-scale knowledgebase pre-training, LLMs had great potential in drug development analysis owing to their holistic understanding of natural language and human topics. LEDAP illustrated its notable competitiveness in comparison with other popular DBA analysis tools. Specifically, even in simple conjunction with classical machine learning methods, LLM-based feature representations consistently enabled satisfactory performance across diverse DBA tasks like binary classification, multiclass classification, and regression. Our findings underpinned the considerable potential of LLMs in drug development research, indicating a catalyst for further progress in related fields.
RESUMO
Ufmylation is implicated in multiple cellular processes, but little is known about its functions and regulation in protein trafficking. Here, we demonstrate that the genetic depletion of core components of the ufmylation cascade, including ubiquitin-fold modifier 1 (UFM1), UFM1 activation enzyme 5, UFM1-specific ligase 1 (UFL1), UFM1-specific protease 2, and UFM1-binding protein 1 (UFBP1) each markedly inhibits the endoplasmic reticulum (ER)-Golgi transport, surface delivery, and recruitment to COPII vesicles of a subset of G protein-coupled receptors (GPCRs) and UFBP1's function partially relies on UFM1 conjugation. We also show that UFBP1 and UFL1 interact with GPCRs and UFBP1 localizes at COPII vesicles coated with specific Sec24 isoforms. Furthermore, the UFBP1/UFL1-binding domain identified in the receptors effectively converts non-GPCR protein transport into the ufmylation-dependent pathway. Collectively, these data reveal important functions for the ufmylation system in GPCR recruitment to COPII vesicles, biosynthetic transport, and sorting at ER via UFBP1 ufmylation and interaction directly.
Assuntos
Vesículas Revestidas pelo Complexo de Proteína do Envoltório , Retículo Endoplasmático , Transporte Proteico , Receptores Acoplados a Proteínas G , Vesículas Revestidas pelo Complexo de Proteína do Envoltório/metabolismo , Retículo Endoplasmático/metabolismo , Receptores Acoplados a Proteínas G/metabolismo , Receptores Acoplados a Proteínas G/genética , Humanos , Complexo de Golgi/metabolismo , Ligação Proteica , Proteínas de Transporte Vesicular/metabolismo , Proteínas de Transporte Vesicular/genética , Células HEK293 , Células HeLa , ProteínasRESUMO
Degradation of target proteins has been considered to be a promising therapeutic approach, but the rational design of compounds for degradation remains a challenge. In this study, we reasonably designed and synthesized only 10 compounds to discover effective CDK4/6 protein degraders. Among the newly synthesized compounds, 7f achieved dual degradation of CDK4/6 protein, with DC50 values of 10.5 and 2.5 nM, respectively. Compound 7f also exhibited inhibitory proliferative activity against Jurkat cells with an IC50 value of 0.18 µM. Furthermore, 7f induced cell apoptosis and G1 phase cell cycle arrest in a dose-dependent manner in Jurkat cells. In conclusion, these findings demonstrate the potential of 7f as a CDK4/6 degrader and a potential therapeutic strategy against cancer, thereby expanding the potential of CDK4/6 dual PROTACs.
Assuntos
Antineoplásicos , Apoptose , Proliferação de Células , Quinase 4 Dependente de Ciclina , Quinase 6 Dependente de Ciclina , Desenho de Fármacos , Humanos , Quinase 4 Dependente de Ciclina/antagonistas & inibidores , Quinase 4 Dependente de Ciclina/metabolismo , Quinase 6 Dependente de Ciclina/antagonistas & inibidores , Quinase 6 Dependente de Ciclina/metabolismo , Antineoplásicos/farmacologia , Antineoplásicos/química , Antineoplásicos/síntese química , Proliferação de Células/efeitos dos fármacos , Apoptose/efeitos dos fármacos , Inibidores de Proteínas Quinases/farmacologia , Inibidores de Proteínas Quinases/química , Inibidores de Proteínas Quinases/síntese química , Células Jurkat , Relação Estrutura-Atividade , Proteólise/efeitos dos fármacos , Estrutura MolecularRESUMO
Single-cell transcriptome sequencing (scRNA-seq) is a high-throughput technique used to study gene expression at the single-cell level. Clustering analysis is a commonly used method in scRNA-seq data analysis, helping researchers identify cell types and uncover interactions between cells. However, the choice of a robust similarity metric in the clustering procedure is still an open challenge due to the complex underlying structures of the data and the inherent noise in data acquisition. Here, we propose a deep clustering method for scRNA-seq data called scRISE (scRNA-seq Iterative Smoothing and self-supervised discriminative Embedding model) to resolve this challenge. The model consists of two main modules: an iterative smoothing module based on graph autoencoders designed to denoise the data and refine the pairwise similarity in turn to gradually incorporate cell structural features and enrich the data information; and a self-supervised discriminative embedding module with adaptive similarity threshold for partitioning samples into correct clusters. Our approach has shown improved quality of data representation and clustering on seventeen scRNA-seq datasets against a number of state-of-the-art deep learning clustering methods. Furthermore, utilizing the scRISE method in biological analysis against the HNSCC dataset has unveiled 62 informative genes, highlighting their potential roles as therapeutic targets and biomarkers.