P53 upregulation by USP7-engaging molecular glues.

Molecular glues are typically small chemical molecules that act at the interface between a target protein and degradation machinery to trigger ternary complex formation. Identifying molecular glues is challenging. There is a scarcity of target-specific upregulating molecular glues, which are highly anticipated for numerous targets, including P53. P53 is degraded in proteasomes through polyubiquitination by specific E3 ligases, whereas deubiquitinases (DUBs) remove polyubiquitination conjugates to counteract these E3 ligases. Thus, small-molecular glues that enhance P53 anchoring to DUBs may stabilize P53 through deubiquitination. Here, using small-molecule microarray-based technology and unbiased screening, we identified three potential molecular glues that may tether P53 to the DUB, USP7, and elevate the P53 level. Among the molecular glues, bromocriptine (BC) is an FDA-approved drug with the most robust effects. BC was further verified to increase P53 stability via the predicted molecular glue mechanism engaging USP7. Consistent with P53 upregulation in cancer cells, BC was shown to inhibit the proliferation of cancer cells in vitro and suppress tumor growth in a xenograft model. In summary, we established a potential screening platform and identified potential molecular glues upregulating P53. Similar strategies could be applied to the identification of other types of molecular glues that may benefit drug discovery and chemical biology studies.

Trenched microwave resonator integrated with porous PDMS for detection and classification of VOCs with enhanced performance.

Microwave resonators combined with polymer absorption layers are widely used in volatile organic compound (VOC) detection based on their variable resonant frequencies. However, the response time is limited due to the polymer's slow volumetric absorption of VOC molecules. By constructing a porous structure in Polydimethylsiloxane (PDMS), resulting in reduced the response time to as short as 71.1%. To mitigate the sensitivity decline caused by the porous PDMS, a trenched-substrate complementary split-ring resonator (CSRR) is proposed for enhancing the interaction between the electromagnetic fields (EMFs) and the porous PDMS with VOCs. The removal of the substrate beneath CSRR's sensing region enhances the effective EMF, increasing frequency and amplitude sensitivities up to 175.5% and 137.8%, respectively. Responses to four common VOCs by the sensor show a maximum sensitivity of 217 Hz/ppm and a minimum limit of detection of 295 ppm. Additionally, resonant parameters and extracted lumped parameters are utilized to establish two decision-tree-based VOC classification models, achieving high accuracies of 98.71% and 99.59%, respectively. And the latter one fully utilizing responses throughout the swept band, proves superior in identifying similar substances. This sensor technology helps promote the sensitive detection and accurate classification of diverse VOCs.

Single-cell sequencing analysis and multiple machine-learning models revealed the cellular crosstalk of dendritic cells and identified FABP5 and KLRB1 as novel biomarkers for psoriasis.

Background: Psoriasis is an immune-mediated disorder influenced by environmental factors on a genetic basis. Despite advancements, challenges persist, including the diminishing efficacy of biologics and small-molecule targeted agents, alongside managing recurrence and psoriasis-related comorbidities. Unraveling the underlying pathogenesis and identifying valuable biomarkers remain pivotal for diagnosing and treating psoriasis. Methods: We employed a series of bioinformatics (including single-cell sequencing data analysis and machine learning techniques) and statistical methods to integrate and analyze multi-level data. We observed the cellular changes in psoriatic skin tissues, screened the key genes Fatty acid binding protein 5 (FABP5) and The killer cell lectin-like receptor B1 (KLRB1), evaluated the efficacy of six widely prescribed drugs on psoriasis treatment in modulating the dendritic cell-associated pathway, and assessed their overall efficacy. Finally, RT-qPCR, immunohistochemistry, and immunofluorescence assays were used to validate. Results: The regulatory influence of dendritic cells (DCs) on T cells through the CD70/CD27 signaling pathway may emerge as a significant facet of the inflammatory response in psoriasis. Notably, FABP5 and KLRB1 exhibited up-regulation and co-localization in psoriatic skin tissues and M5-induced HaCaT cells, serving as potential biomarkers influencing psoriasis development. Conclusion: Our study analyzed the impact of DC-T cell crosstalk in psoriasis, elucidated the characterization of two biomarkers, FABP5 and KLRB1, in psoriasis, and highlighted the promise and value of tofacitinib in psoriasis therapy targeting DCs.

Exhaustive Exploitation of Nature-Inspired Computation for Cancer Screening in an Ensemble Manner.

Accurate screening of cancer types is crucial for effective cancer detection and precise treatment selection. However, the association between gene expression profiles and tumors is often limited to a small number of biomarker genes. While computational methods using nature-inspired algorithms have shown promise in selecting predictive genes, existing techniques are limited by inefficient search and poor generalization across diverse datasets. This study presents a framework termed Evolutionary Optimized Diverse Ensemble Learning (EODE) to improve ensemble learning for cancer classification from gene expression data. The EODE methodology combines an intelligent grey wolf optimization algorithm for selective feature space reduction, guided random injection modeling for ensemble diversity enhancement, and subset model optimization for synergistic classifier combinations. Extensive experiments were conducted across 35 gene expression benchmark datasets encompassing varied cancer types. Results demonstrated that EODE obtained significantly improved screening accuracy over individual and conventionally aggregated models. The integrated optimization of advanced feature selection, directed specialized modeling, and cooperative classifier ensembles helps address key challenges in current nature-inspired approaches. This provides an effective framework for robust and generalized ensemble learning with gene expression biomarkers. Specifically, we have opened EODE source code on Github at https://github.com/wangxb96/EODE.

Trends in the Prevalence of Common Retinal and Optic Nerve Diseases in China: An Artificial Intelligence Based National Screening.

Purpose: Retinal and optic nerve diseases have become the primary cause of irreversible vision loss and blindness. However, there is still a lack of thorough evaluation regarding their prevalence in China. Methods: This artificial intelligence-based national screening study applied a previously developed deep learning algorithm, named the Retinal Artificial Intelligence Diagnosis System (RAIDS). De-identified personal medical records from January 2019 to December 2021 were extracted from 65 examination centers in 19 provinces of China. Crude prevalence and age-sex-adjusted prevalence were calculated by mapping to the standard population in the seventh national census. Results: In 2021, adjusted referral possible glaucoma (63.29, 95% confidence interval [CI] = 57.12-68.90 cases per 1000), epiretinal macular membrane (21.84, 95% CI = 15.64-29.22), age-related macular degeneration (13.93, 95% CI = 11.09-17.17), and diabetic retinopathy (11.33, 95% CI = 8.89-13.77) ranked the highest among 10 diseases. Female participants had significantly higher adjusted prevalence of pathologic myopia, yet a lower adjusted prevalence of diabetic retinopathy, referral possible glaucoma, and hypertensive retinopathy than male participants. From 2019 to 2021, the adjusted prevalence of retinal vein occlusion (0.99, 95% CI = 0.73-1.26 to 1.88, 95% CI = 1.42-2.44), macular hole (0.59, 95% CI = 0.41-0.82 to 1.12, 95% CI = 0.76-1.51), and hypertensive retinopathy (0.53, 95% CI = 0.40-0.67 to 0.77, 95% CI = 0.60-0.95) significantly increased. The prevalence of diabetic retinopathy in participants under 50 years old significant increased. Conclusions: Retinal and optic nerve diseases are an important public health concern in China. Further well-conceived epidemiological studies are required to validate the observed increased prevalence of diabetic retinopathy, hypertensive retinopathy, retinal vein occlusion, and macular hole nationwide. Translational Relevance: This artificial intelligence system can be a potential tool to monitor the prevalence of major retinal and optic nerve diseases over a wide geographic area.

A Highly Sensitive Deep-Sea Hydrodynamic Pressure Sensor Inspired by Fish Lateral Line.

Hydrodynamic pressure sensors offer an auxiliary approach for ocean exploration by unmanned underwater vehicles (UUVs). However, existing hydrodynamic pressure sensors often lack the ability to monitor subtle hydrodynamic stimuli in deep-sea environments. In this study, we present the development of a deep-sea hydrodynamic pressure sensor (DSHPS) capable of operating over a wide range of water depths while maintaining exceptional hydrodynamic sensing performance. The DSHPS device was systematically optimized by considering factors such as piezoelectric polyvinylidene fluoride-trifluoroethylene/barium titanate [P(VDF-TrFE)/BTO] nanofibers, electrode configurations, sensing element dimensions, integrated circuits, and packaging strategies. The optimized DSHPS exhibited a remarkable pressure gradient response, achieving a minimum pressure difference detection capability of approximately 0.11 Pa. Additionally, the DSHPS demonstrated outstanding performance in the spatial positioning of dipole sources, which was elucidated through theoretical charge modeling and fluid-structure interaction (FSI) simulations. Furthermore, the integration of a high Young's modulus packaging strategy inspired by fish skull morphology ensured reliable sensing capabilities of the DSHPS even at depths of 1000 m in the deep sea. The DSHPS also exhibited consistent and reproducible positioning performance for subtle hydrodynamic stimulus sources across this wide range of water depths. We envision that the development of the DSHPS not only enhances our understanding of the evolutionary aspects of deep-sea canal lateral lines but also paves the way for the advancement of artificial hydrodynamic pressure sensors.

A three-dimensional liquid diode for soft, integrated permeable electronics.

Wearable electronics with great breathability enable a comfortable wearing experience and facilitate continuous biosignal monitoring over extended periods1-3. However, current research on permeable electronics is predominantly at the stage of electrode and substrate development, which is far behind practical applications with comprehensive integration with diverse electronic components (for example, circuitry, electronics, encapsulation)4-8. Achieving permeability and multifunctionality in a singular, integrated wearable electronic system remains a formidable challenge. Here we present a general strategy for integrated moisture-permeable wearable electronics based on three-dimensional liquid diode (3D LD) configurations. By constructing spatially heterogeneous wettability, the 3D LD unidirectionally self-pumps the sweat from the skin to the outlet at a maximum flow rate of 11.6 ml cm-2 min-1, 4,000 times greater than the physiological sweat rate during exercise, presenting exceptional skin-friendliness, user comfort and stable signal-reading behaviour even under sweating conditions. A detachable design incorporating a replaceable vapour/sweat-discharging substrate enables the reuse of soft circuitry/electronics, increasing its sustainability and cost-effectiveness. We demonstrated this fundamental technology in both advanced skin-integrated electronics and textile-integrated electronics, highlighting its potential for scalable, user-friendly wearable devices.

Anti-Aging in Caenorhabditis elegans of Polysaccharides from Polygonatum cyrtonema Hua.

Polygonatum cyrtonema Hua, the dried rhizome of Polygonum multiflorum from the Liliaceae family, is a widely used medicinal herb with a long history of application. Its main active ingredients are polysaccharides, which have been demonstrated in contemporary studies to effectively delay the aging process. In the present study, homogeneous polysaccharide (PCP-1) was obtained after the purification and isolation of polysaccharides from Polygonatum cyrtonema Hua (PCP). The anti-aging activities of both were compared, and the possible mechanism of action for exerting anti-aging activity was explored using Caenorhabditis elegans (C. elegans). Research has indicated that PCP and PCP-1 exhibit potent anti-oxidant and anti-aging properties. Of particular note is that PCP-1 acts better than PCP. The two were able to prolong the lifespan of nematodes, improve the stress resistance of nematodes, reduce the accumulation of lipofuscin in the intestine, decrease the content of ROS and MDA in the body, increase the activity of the antioxidant enzymes SOD and CAT, promote the nuclear translocation of DAF-16, down-regulate the mRNA levels of the age-1 and daf-2 genes of the IIS pathway in nematodes, and up-regulate the expression of the daf-16, skn-1, sod-3, and hsp-16.2 genes. Based on the aforementioned findings, it is possible that the mechanism by which PCP and PCP-1 exert anti-aging effects may be through negative regulation of the IIS pathway, activation of the transcription factor DAF-16/FOXO, and enhancement of oxidative defenses and stress resistance in nematodes. Overall, the present study illustrated the great potential of polysaccharides from Polygonatum cyrtonema Hua in anti-aging and antioxidant activities. Specifically, PCP-1 demonstrated superior characteristics, which provides a reference for the future development of Polygonatum cyrtonema Hua polysaccharides.

SPRR1B is Related to the Immune Microenvironment and Can Be Used as a Biomarker for the Diagnosis of Psoriasis.

Background: Psoriasis, a chronic inflammatory disorder with an unknown cause, significantly impacts the physical and psychological well-being of patients. However, current biomarkers related to psoriasis lack clinical specificity, sensitivity, and predictive ability. Methods: In this study, we collected skin lesion tissues from 20 psoriasis patients and 20 normal skin samples. Additionally, we obtained four datasets from the GEO database, which included human psoriasis and healthy specimens. We utilized SVM-RFE analysis and the LASSO regression model to identify potential biomarkers. Furthermore, we examined the composition of immune cell types in psoriasis and their correlation with specific genes. Results: Our investigation revealed 57 differentially expressed genes (DEGs), and we identified significantly enriched pathways through KEGG pathway analysis. The results of machine learning and WGCNA suggested that LCE3D and SPRR1B could potentially be used as marker genes for diagnosing psoriasis. RT-PCR and immunohistochemical detection confirmed the abnormally high expression of the SPRR1B gene in psoriasis. Analysis of immune cell infiltration revealed a strong positive correlation between SPRR1B and Macrophages M0 and T cells follicular helper, while showing the strongest negative correlation with resting Mast cells. In addition, we found that silencing SPRR1B in IFN-γ-treated HaCat cells could significantly reduce the increase in IL-17, IL-22, KRT6, and KRT16 caused by IFN-γ. Conclusion: These findings suggest that SPRR1B may have a significant role in the pathogenesis of psoriasis and could be employed as a novel immunomarker for its development.

Comparison Between Drug-Coated Balloon and Stents in Large De Novo Coronary Artery Disease: A Systematic Review and Meta-Analysis of RCT Data.

PURPOSE: Although a number of studies involving small-vessel de novo coronary disease showed clinical benefits of drug-coated balloons (DCB), the role of DCB in large vessel lesions is still unclear. METHODS: We searched main electronic databases for randomized controlled trials (RCTs) comparing DCB with stents for large vessel de novo coronary artery disease. The primary endpoint was major cardiovascular adverse events (MACE), composite cardiovascular death (CD), myocardial infarction (MI), or target lesion revascularization (TLR). RESULTS: This study included 7 RCTs with 770 participants. DCB were associated with a marked risk reduction in MACE [Risk Ratio (RR): 0.48; 95% confidence interval [CI]: 0.24 to 0.97; P = 0.04], TLR (RR: 0.53; 95% CI: 0.25 to 1.14; P = 0.10), and late lumen loss [standard mean difference (SMD): -0.57; 95% CI: -1.09 to -0.05; P = 0.03] as compared with stents. There is no significant difference in MI (RR: 0.58; 95% CI: 0.21 to 1.54; P = 0.27), CD (RR: 0.33; 95% CI: 0.06 to 1.78; P = 0.19), and minimal lumen diameter (SMD: -0.34; 95% CI: -0.72 to 0.05; P = 0.08) between groups. In subgroup analyses, the risk reduction of MACE persisted in patients with chronic coronary syndrome (RR: 0.25; 95% CI: 0.07 to 0.89; P = 0.03), and patients receiving DCB vs. bare metal stent (RR: 0.19; 95% CI: 0.05 to 0.73; P = 0.01). In addition, there was no significant difference between the DCB group and the drug eluting stent group for MACE (RR: 0.69; 95% CI: 0.30 to 1.60; P = 0.38). CONCLUSION: DCB may be an effective therapeutic option in patients with large vessel de novo coronary artery disease.

Hydrophobic MOF/PDMS-Based QCM Sensors for VOCs Identification and Quantitative Detection in High-Humidity Environments.

Metal-organic frameworks (MOFs) have great potential in quartz crystal microbalance (QCM) platforms for volatile organic compound (VOCs) detection and recognition due to their unique properties. However, the MOFs' hydrophilicity degrades performance in high-humidity environments, limiting reliable VOC sensing in complex environments. Herein, we propose a novel VOC virtual sensor array (VSA) using a single QCM sensor with an adsorption layer composed of MIL-101(Cr) MOF and polydimethylsiloxane (PDMS), realizing stable sensing and accurate identification for different VOCs under various relative humidity (RH) conditions. The hydrophobic PDMS layer improves the moisture resistance of the sensor to 4 and 14 times in terms of shifts in resonant frequency and scattering parameters, respectively. In addition, performance is maintained over 2 days of water treatment, demonstrating superior water resistance. The highest sensitivity of 2.68 mdB ppm-1 is achieved for isopropanol detection, with the lowest limit of detection of 20.06 ppm for acetone. Combining resonant signals and lumped parameters, the proposed VSA technique effectively discriminates four VOCs (ethanol, 2-propanol, acetone, and acetonitrile) with a high accuracy of 95.3% under both 60% and 90% RH backgrounds. The studies provide a promising solution for reliable low-concentration VOC detection using QCM sensors in high-humidity environments such as underground spaces.

Targeting METTL3 reprograms the tumor microenvironment to improve cancer immunotherapy.

The tumor microenvironment (TME) is a heterogeneous ecosystem containing cancer cells, immune cells, stromal cells, cytokines, and chemokines which together govern tumor progression and response to immunotherapies. Methyltransferase-like 3 (METTL3), a core catalytic subunit for RNA N6-methyladenosine (m6A) modification, plays a crucial role in regulating various physiological and pathological processes. Whether and how METTL3 regulates the TME and anti-tumor immunity in non-small-cell lung cancer (NSCLC) remain poorly understood. Here, we report that METTL3 elevates expression of pro-tumorigenic chemokines including CXCL1, CXCL5, and CCL20, and destabilizes PD-L1 mRNA in an m6A-dependent manner, thereby shaping a non-inflamed TME. Thus, inhibiting METTL3 reprograms a more inflamed TME that renders anti-PD-1 therapy more effective in several murine lung tumor models. Clinically, NSCLC patients who exhibit low-METTL3 expression have a better prognosis when receiving anti-PD-1 therapy. Collectively, our study highlights targeting METTL3 as a promising strategy to improve immunotherapy in NSCLC patients.

Nearly optimal fault-tolerant constrained tracking for multi-axis servo system via practical terminal sliding mode and adaptive dynamic programming.

In this paper, a nearly optimal tracking control is proposed for n-links robotic manipulators subject to parameter uncertainties, time-profile failures, and input saturation constraints. Firstly, the practical terminal sliding-mode (PTSM) manifold with a linear additional term is proposed to combine the system states related to joint rotation, such that the controlled states quickly fall into a tiny neighborhood of the equilibrium once they reach the PTSM manifold. Secondly, a nearly optimal sliding-mode reaching law is designed by using the adaptive dynamic programming (ADP) technique. Benefiting from a non-quadratic positive defined mapping of the proposed performance index, which relates to the derivative of the sliding-mode function, reduced-order system dynamics can be constrained to a desired region. For the bounded actuator fault caused by various inducements such as the power supply fluctuation and the wear of parts, a radial basis function neural network (RBFNN) is introduced to compensate for this, and the input saturation constraints of the controlled plant are also compensated at the same time. Innovatively, the node weights of RBFNN are updated by the critic network of the ADP framework, such that the integrity of the proposed control strategy is improved. Simulations verify the main conclusions.

Asymmetric Synthesis of Arboduridine.

The first asymmetric total synthesis of the monoterpenoid indole alkaloid arboduridine has been accomplished. The tricyclic A/B/D ring system was constructed by an enantioselective Michael reaction followed by intramolecular nucleophilic addition. Intramolecular α-amination of a ketone forged the piperidine ring, while a Horner-Wadsworth-Emmons (HWE) reaction was used to form the pyrrolidine ring. A reduction cyclization cascade led to formation of the tetrahydrofuran ring.

Severe photochemical pollution formation associated with strong HONO emissions from dew and guttation evaporation.

The unknown daytime source of HONO has been extensively investigated due to unexplained atmospheric oxidation capacity and current modelling bias, especially during cold seasons. In this study, abrupt morning increases in atmospheric HONO at a rural site in the North China Plain (NCP) were observed almost on daily basis, which were closely linked to simultaneous rises in atmospheric water vapor content and NH3 concentrations. Dew and guttation water formation was frequently observed on wheat leaves, from which water samples were taken and chemically analyzed for the first time. Results confirmed that such natural processes likely governed the daily nighttime deposition and daytime release of HONO and NH3, which have not been considered in the numerous HONO budget studies investigating its large missing daytime source in the NCP. The dissolved HONO and NH3 in leaf surface water droplets reached 1.4 and 23 mg L-1 during the morning on average, resulting in averaged atmospheric HONO and NH3 increases of 0.89 ± 0.61 and 43.7 ± 29.3 ppb during morning hours, with relative increases of 186 ± 212 % and 233 ± 252 %, respectively. The high atmospheric oxidation capacity contained within HONO was stored in near surface liquid water (such as dew, guttation and soil surface water) during nighttime, which prevented its atmospheric dispersion after sunset and protected it from photodissociation during early morning hours. HONO was released in a blast during later hours with stronger solar radiation, which triggered and then accelerated daytime photochemistry through the rapid photolysis of HONO and subsequent OH production, especially under high RH conditions, forming severe secondary gaseous and particulate pollution. Results of this study demonstrate that global ecosystems might play significant roles in atmospheric photochemistry through nighttime dew formation and guttation processes.

A multi-view convolutional neural network method combining attention mechanism for diagnosing autism spectrum disorder.

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition whose current psychiatric diagnostic process is subjective and behavior-based. In contrast, functional magnetic resonance imaging (fMRI) can objectively measure brain activity and is useful for identifying brain disorders. However, the ASD diagnostic models employed to date have not reached satisfactory levels of accuracy. This study proposes the use of MAACNN, a method that utilizes multi-view convolutional neural networks (CNNs) in conjunction with attention mechanisms for identifying ASD in multi-scale fMRI. The proposed algorithm effectively combines unsupervised and supervised learning. In the initial stage, we employ stacked denoising autoencoders, an unsupervised learning method for feature extraction, which provides different nodes to adapt to multi-scale data. In the subsequent stage, we perform supervised learning by employing multi-view CNNs for classification and obtain the final results. Finally, multi-scale data fusion is achieved by using the attention fusion mechanism. The ABIDE dataset is used to evaluate the model we proposed., and the experimental results show that MAACNN achieves superior performance with 75.12% accuracy and 0.79 AUC on ABIDE-I, and 72.88% accuracy and 0.76 AUC on ABIDE-II. The proposed method significantly contributes to the clinical diagnosis of ASD.

Serum lymphocytes and cytokines: diagnostic value and influence on the immune status in patients with pulmonary tuberculosis.

OBJECTIVE: To determine the absolute number of serum T lymphocytes and cytokine levels and the characteristics of patients with active pulmonary tuberculosis and to assess their effect on the immune status of these patients and their diagnostic and predictive value for tuberculosis. METHODS: We included 1,069 patients with active tuberculosis, 51 patients with latent tuberculosis infection, and 600 health individuals. Absolute serum T-lymphocyte counts and cytokine levels were quantified. RESULTS: T lymphocytes were significantly reduced in patients with active tuberculosis when compared with healthy individuals. The immune function of patients gradually decreased with age and was stronger in female patients than in males. Th1 cells expressed higher levels of cytokines than did Th2 cells. Logistic regression analysis showed that reduced CD3+ T, CD8+ T, and NK cell counts, as well as reduced IL-4 and IFN-g expression, were independent influencing factors for active tuberculosis. ROC analysis showed that the sensitivity and specificity of absolute CD3+ T and CD8+ T lymphocyte counts and combined factors were significantly higher than were those of IL-4 and IFN-g for diagnosing active tuberculosis. CONCLUSIONS: Serum T-lymphocyte counts and cytokine levels can assess the immune status of tuberculosis patients; they are also useful biomarkers for predicting and diagnosing tuberculosis.

A deformability-based biochip for precise label-free stratification of metastatic subtypes using deep learning.

Cellular deformability is a promising biomarker for evaluating the physiological state of cells in medical applications. Microfluidics has emerged as a powerful technique for measuring cellular deformability. However, existing microfluidic-based assays for measuring cellular deformability rely heavily on image analysis, which can limit their scalability for high-throughput applications. Here, we develop a parallel constriction-based microfluidic flow cytometry device and an integrated computational framework (ATMQcD). The ATMQcD framework includes automatic training set generation, multiple object tracking, segmentation, and cellular deformability quantification. The system was validated using cancer cell lines of varying metastatic potential, achieving a classification accuracy of 92.4% for invasiveness assessment and stratifying cancer cells before and after hypoxia treatment. The ATMQcD system also demonstrated excellent performance in distinguishing cancer cells from leukocytes (accuracy = 89.5%). We developed a mechanical model based on power-law rheology to quantify stiffness, which was fitted with measured data directly. The model evaluated metastatic potentials for multiple cancer types and mixed cell populations, even under real-world clinical conditions. Our study presents a highly robust and transferable computational framework for multiobject tracking and deformation measurement tasks in microfluidics. We believe that this platform has the potential to pave the way for high-throughput analysis in clinical applications, providing a powerful tool for evaluating cellular deformability and assessing the physiological state of cells.

THPLM: a sequence-based deep learning framework for protein stability changes prediction upon point variations using pretrained protein language model.

MOTIVATION: Quantitative determination of protein thermodynamic stability is a critical step in protein and drug design. Reliable prediction of protein stability changes caused by point variations contributes to developing-related fields. Over the past decades, dozens of structure-based and sequence-based methods have been proposed, showing good prediction performance. Despite the impressive progress, it is necessary to explore wild-type and variant protein representations to address the problem of how to represent the protein stability change in view of global sequence. With the development of structure prediction using learning-based methods, protein language models (PLMs) have shown accurate and high-quality predictions of protein structure. Because PLM captures the atomic-level structural information, it can help to understand how single-point variations cause functional changes. RESULTS: Here, we proposed THPLM, a sequence-based deep learning model for stability change prediction using Meta's ESM-2. With ESM-2 and a simple convolutional neural network, THPLM achieved comparable or even better performance than most methods, including sequence-based and structure-based methods. Furthermore, the experimental results indicate that the PLM's ability to generate representations of sequence can effectively improve the ability of protein function prediction. AVAILABILITY AND IMPLEMENTATION: The source code of THPLM and the testing data can be accessible through the following links: https://github.com/FPPGroup/THPLM.