Search Results (21)

Proteins are dynamic macromolecules whose functions are intricately linked to their structural flexibility. Recent breakthroughs in deep learning have enabled accurate prediction of static protein structures. However, understanding protein function is more complex. It often requires access to a diverse ensemble of conformations. Traditional sampling techniques exist to help with this. These include molecular dynamics and Monte Carlo simulations. These techniques can explore conformational landscapes. However, they have limitations as they are often limited by high computational cost and suffer from slow convergence. In response, deep generative models (DGMs) have emerged as a powerful alternative for efficient and scalable protein conformation sampling. Leveraging architectures such as variational autoencoders, normalizing flows, generative adversarial networks, and diffusion models, DGMs can learn complex, high-dimensional distributions over protein conformations directly from data. This survey on generative models for protein conformation sampling provides a comprehensive overview of recent advances in this emerging field. We categorize existing models based on generative architecture, structural representation, and target tasks. We also discuss key datasets, evaluation metrics, limitations, and opportunities for integrating physics-based knowledge with data-driven models. By bridging machine learning and structural biology, DGMs are poised to transform our ability to model, design, and understand dynamic protein behavior. Full article

Ischemic stroke is a primary cause of cerebrovascular diseases and continues to be one of the leading causes of death and disability among patients worldwide. Pathological processes caused by vascular damage due to stroke occur in a time-dependent manner and are classified into three categories: acute, subacute, and chronic. Current treatments for ischemic stroke are limited to effectiveness in the early stages. In this study, we investigated the therapeutic effect of an oriental medicine, Gosha-jinki-gan (TJ107), on improving chronic ischemic stroke using the mouse model with middle cerebral artery occlusion (MCAO). The changes in the intracerebral inflammatory response (macrophages (F4/80), TLR2•4, IL-23, IL-17, TNF-α, and IL-1β) were examined using real-time RT-PCR. The MCAO mice showed the increased expression of glial fibrillary acidic protein (GFAP) and of F4/80, TLR2, TLR4, IL-1β, TNF-α, and IL-17 in the brain tissue from the MCAO region. This suggests that they contribute to the expansion of the ischemic stroke infarct area and to the worsening of the neurological symptoms of the MCAO mice in the chronic phase. On the other hand, the administration of TJ107 was proven to reduce the infarct area, with decreased GFAP expression, suppressed macrophage infiltration in the brain, and reduced TNF-α, IL-1β, and IL-17 production compared with the MCAO mice. This study first demonstrated Gosha-jinki-gan’s therapeutic effects on the chronic ischemic stroke. Full article

Weaning stress damages the intestines and disrupts the intestinal barrier in piglets, which significantly impacts the pig farming industry’s economy. We aimed to examine the effects of ShenQiGan extract (CAG) on intestinal barrier function and explore the underlying molecular mechanisms in stress-challenged weaned piglets. The experimental design involved 80 weaned piglets aged 28 days (with an average body weight of 7.78 ± 0.074 kg) that were randomly allocated into four groups: Control, LCAG (0.1% CAG), MCAG (0.5% CAG), and HCAG (1.0% CAG). After a 28-day trial period, the growth performance and incidence of diarrhea in piglets were evaluated. CAG increased the average daily gain of weaned piglets, reduced the feed-to-gain ratio, and decreased the incidence of diarrhea. It significantly lowered serum inflammatory cytokine levels while elevating immunoglobulin levels. The supplement notably enhanced concentrations of acetic acid, propionic acid, butyric acid, and isobutyric acid. Furthermore, CAG demonstrated intestinal morphology restoration and upregulation of tight junction proteins and MUC2 protein expression in jejunum. At the mRNA level, it significantly upregulated the expression of Occludin, Claudin1, and MUC2 genes. CAG improves growth performance and mitigates diarrhea in weaned piglets by enhancing intestinal barrier integrity, modulating systemic inflammatory responses, elevating immunoglobulin levels, and promoting short-chain fatty acids (SCFAs) production in the cecum. Full article

The prognosis of advanced (UICC IIb-IV) primary colorectal cancer (pCRC) remains poor. More effective targeted therapies are needed. Heat shock protein 90 alpha/beta (Hsp90α/β) expression was immunohistologically quantified in 89 pCRCs and multivariately correlated with survival. Pimitespib (Pim, TAS-116), a Hsp90α/β-specific inhibitor, was tested in pCRC cell lines and patient-derived cancer spheroids (PDCS) and referenced to the pan-Hsp90 inhibitor ganetespib (Gan, STA-9090) and standard-of-care therapies. A total of 26.97% pCRCs showed strong tumoral Hsp90α/β expression (Hsp90α/β > 40%), which correlated with reduced PFS (HR: 3.785, 95%CI: 1.578–9.078, p = 0.003) and OS (HR: 3.502, 95%CI: 1.292–9.494, p = 0.014). Co-expression of Hsp90α/β > 40% with its clients BRAF-V600E and Her2/neu aggravated the prognosis (BRAF-V600E mutated: PFS, p = 0.002; OS, p = 0.012; Her2/neu score3: PFS, p = 0.029). The prognostic cut-off Hsp90α/β > 40% was also a predictor for response to Pim-based therapy. Pim efficacy was increased in combination with 5-FU, 5-FU + oxaliplatin, and 5-FU + irinotecan (all p < 0.001). Pim induced sensitization to all chemotherapies in HT-29 (p < 0.001), Caco-2 (p < 0.01), and HCT116 (p < 0.05) cells. Pim combined with encorafenib in HT-29 and with trastuzumab in Caco-2 cells was most effective in dual-target inhibition approaches (HT-29: p < 0.005; Caco-2: p < 0.05). The anti-cancer effect and chemosensitization of Pim-based therapy were prospectively confirmed in PDCS directly generated from Hsp90α/β > 40% pCRCs. Protein profiling combined with functional drug testing stratifies Hsp90α/β > 40% pCRC patients diagnosed with UICC IIb-IV for effective Pim-based therapy. Full article

Deep learning (DL) has become a core component of modern artificial intelligence (AI), driving significant advancements across diverse fields by facilitating the analysis of complex systems, from protein folding in biology to molecular discovery in chemistry and particle interactions in physics. However, the field of deep learning is constantly evolving, with recent innovations in both architectures and applications. Therefore, this paper provides a comprehensive review of recent DL advances, covering the evolution and applications of foundational models like convolutional neural networks (CNNs) and Recurrent Neural Networks (RNNs), as well as recent architectures such as transformers, generative adversarial networks (GANs), capsule networks, and graph neural networks (GNNs). Additionally, the paper discusses novel training techniques, including self-supervised learning, federated learning, and deep reinforcement learning, which further enhance the capabilities of deep learning models. By synthesizing recent developments and identifying current challenges, this paper provides insights into the state of the art and future directions of DL research, offering valuable guidance for both researchers and industry experts. Full article

The complexities inherent in drug development are multi-faceted and often hamper accuracy, speed and efficiency, thereby limiting success. This review explores how recent developments in machine learning (ML) are significantly impacting target-based drug discovery, particularly in small-molecule approaches. The Simplified Molecular Input Line Entry System (SMILES), which translates a chemical compound’s three-dimensional structure into a string of symbols, is now widely used in drug design, mining, and repurposing. Utilizing ML and natural language processing techniques, SMILES has revolutionized lead identification, high-throughput screening and virtual screening. ML models enhance the accuracy of predicting binding affinity and selectivity, reducing the need for extensive experimental screening. Additionally, deep learning, with its strengths in analyzing spatial and sequential data through convolutional neural networks (CNNs) and recurrent neural networks (RNNs), shows promise for virtual screening, target identification, and de novo drug design. Fragment-based approaches also benefit from ML algorithms and techniques like generative adversarial networks (GANs), which predict fragment properties and binding affinities, aiding in hit selection and design optimization. Structure-based drug design, which relies on high-resolution protein structures, leverages ML models for accurate predictions of binding interactions. While challenges such as interpretability and data quality remain, ML’s transformative impact accelerates target-based drug discovery, increasing efficiency and innovation. Its potential to deliver new and improved treatments for various diseases is significant. Full article

Protamines play a critical role in DNA compaction and stabilization in sperm cells, significantly influencing male fertility and various biotechnological applications. Traditionally, identifying these proteins is a challenging and time-consuming process due to their species-specific variability and complexity. Leveraging advancements in computational biology, we present PROTA, a novel tool that combines machine learning (ML) and deep learning (DL) techniques to predict protamines with high accuracy. For the first time, we integrate Generative Adversarial Networks (GANs) with supervised learning methods to enhance the accuracy and generalizability of protamine prediction. Our methodology evaluated multiple ML models, including Light Gradient-Boosting Machine (LIGHTGBM), Multilayer Perceptron (MLP), Random Forest (RF), eXtreme Gradient Boosting (XGBOOST), k-Nearest Neighbors (KNN), Logistic Regression (LR), Naive Bayes (NB), and Radial Basis Function-Support Vector Machine (RBF-SVM). During ten-fold cross-validation on our training dataset, the MLP model with GAN-augmented data demonstrated superior performance metrics: 0.997 accuracy, 0.997 F1 score, 0.998 precision, 0.997 sensitivity, and 1.0 AUC. In the independent testing phase, this model achieved 0.999 accuracy, 0.999 F1 score, 1.0 precision, 0.999 sensitivity, and 1.0 AUC. These results establish PROTA, accessible via a user-friendly web application. We anticipate that PROTA will be a crucial resource for researchers, enabling the rapid and reliable prediction of protamines, thereby advancing our understanding of their roles in reproductive biology, biotechnology, and medicine. Full article

A gallium nitride (GaN) semiconductor is one of the most promising materials integrated into biomedical devices to play the roles of connecting, monitoring, and manipulating the activity of biological components, due to its excellent photoelectric properties, chemical stability, and biocompatibility. In this work, it was found that the photogenerated free charge carriers of the GaN substrate, as an exogenous stimulus, served to promote neural stem cells (NSCs) to differentiate into neurons. This was observed through the systematic investigation of the effect of the persistent photoconductivity (PPC) of GaN on the differentiation of primary NSCs from the embryonic rat cerebral cortex. NSCs were directly cultured on the GaN surface with and without ultraviolet (UV) irradiation, with a control sample consisting of tissue culture polystyrene (TCPS) in the presence of fetal bovine serum (FBS) medium. Through optical microscopy, the morphology showed a greater number of neurons with the branching structures of axons and dendrites on GaN with UV irradiation. The immunocytochemical results demonstrated that GaN with UV irradiation could promote the NSCs to differentiate into neurons. Western blot analysis showed that GaN with UV irradiation significantly upregulated the expression of two neuron-related markers, βIII-tubulin (Tuj-1) and microtubule-associated protein 2 (MAP-2), suggesting that neurite formation and the proliferation of NSCs during differentiation were enhanced by GaN with UV irradiation. Finally, the results of the Kelvin probe force microscope (KPFM) experiments showed that the NSCs cultured on GaN with UV irradiation displayed about 50 mV higher potential than those cultured on GaN without irradiation. The increase in cell membrane potential may have been due to the larger number of photogenerated free charges on the GaN surface with UV irradiation. These results could benefit topical research and the application of GaN as a biomedical material integrated into neural interface systems or other bioelectronic devices. Full article

Oil-Gan is the fruit of the genus Phyllanthus emblica L. The fruits have excellent effects on health care and development values. There are many methods for the management of diabetic nephropathy (DN). However, there is a lack of effective drugs for treating DN throughout the disease course. The primary aim of this study was to examine the protective effects (including analyses of urine and blood, and inflammatory cytokine levels) and mechanisms of the ethyl acetate extract of P. emblica (EPE) on db/db mice, an animal model of diabetic nephropathy; the secondary aim was to examine the expression levels of p- protein kinase Cα (PKCα)/t-PKCα in the kidney and its downregulation of vascular endothelial growth factor (VEGF) and fibrosis gene transforming growth factor-β1 (TGF-β1) by Western blot analyses. Eight db/m mice were used as the control group. Forty db/db mice were randomly divided into five groups. Treatments included a vehicle, EPE1, EPE2, EPE3 (at doses of 100, 200, or 400 mg/kg EPE), or the comparative drug aminoguanidine for 8 weeks. After 8 weeks of treatment, the administration of EPE to db/db mice effectively controlled hyperglycemia and hyperinsulinemia by markedly lowering blood glucose, insulin, and glycosylated HbA1c levels. The administration of EPE to db/db mice decreased the levels of BUN and creatinine both in blood and urine and reduced urinary albumin excretion and the albumin creatine ratio (UACR) in urine. Moreover, EPE treatment decreased the blood levels of inflammatory cytokines, including kidney injury molecule-1 (KIM-1), C-reactive protein (CRP), and NLR family pyrin domain containing 3 (NLRP3). Our findings showed that EPE not only had antihyperglycemic effects but also improved renal function in db/db mice. A histological examination of the kidney by immunohistochemistry indicated that EPE can improve kidney function by ameliorating glomerular morphological damage following glomerular injury; alleviating proteinuria by upregulating the expression of nephrin, a biomarker of early glomerular damage; and inhibiting glomerular expansion and tubular fibrosis. Moreover, the administration of EPE to db/db mice increased the expression levels of p- PKCα/t-PKCα but decreased the expression levels of VEGF and renal fibrosis biomarkers (TGF-β1, collagen IV, p-Smad2, p-Smad3, and Smad4), as shown by Western blot analyses. These results implied that EPE as a supplement has a protective effect against renal dysfunction through the amelioration of insulin resistance as well as the suppression of nephritis and fibrosis in a DN model. Full article

Fructose-1,6-bisphosphate aldolase (FBA) is an important enzyme involved in central carbon metabolism (CCM) with promising industrial applications. Artificial intelligence models like generative adversarial networks (GANs) can design novel sequences that differ from natural ones. To expand the sequence space of FBA, we applied the generative adversarial network (ProteinGAN) model for the de novo design of FBA in this study. First, we corroborated the viability of the ProteinGAN model through replicating the generation of functional MDH variants. The model was then applied to the design of class II FBA. Computational analysis showed that the model successfully captured features of natural class II FBA sequences while expanding sequence diversity. Experimental results validated soluble expression and activity for the generated FBAs. Among the 20 generated FBA sequences (identity ranging from 85% to 99% with the closest natural FBA sequences), 4 were successfully expressed as soluble proteins in E. coli, and 2 of these 4 were functional. We further proposed a filter based on sequence identity to the endogenous FBA of E. coli and reselected 10 sequences (sequence identity ranging from 85% to 95%). Among them, six were successfully expressed as soluble proteins, and five of these six were functional—a significant improvement compared to the previous results. Furthermore, one generated FBA exhibited activity that was 1.69fold the control FBA. This study demonstrates that enzyme design with GANs can generate functional protein variants with enhanced performance and unique sequences. Full article

Giant axonal neuropathy (GAN) is a pediatric, hereditary, neurodegenerative disorder that affects both the central and peripheral nervous systems. It is caused by mutations in the GAN gene, which codes for the gigaxonin protein. Gigaxonin plays a role in intermediate filament (IF) turnover hence loss of function of this protein leads to IF aggregates in various types of cells. These aggregates can lead to abnormal cellular function that manifests as a diverse set of symptoms in persons with GAN including nerve degeneration, cognitive issues, skin diseases, vision loss, and muscle weakness. GAN has no cure at this time. Currently, an adeno-associated virus (AAV) 9-mediated gene replacement therapy is being tested in a phase I clinical trial for the treatment of GAN. This review paper aims to provide an overview of giant axonal neuropathy and the current efforts at developing a treatment for this devastating disease. Full article

Licorice (Gan-Cao, licorice) is a natural antioxidant and roasted licorice is the most common processing specification used in traditional Chinese medicine prescriptions. Traditional Chinese medicine theory deems that the honey-roasting process can promote the efficacy of licorice, including tonifying the spleen and augmenting “Qi” (energy). The antioxidant activity and mechanisms underlying roasted licorice have not yet been reported. In this study, we found that roasted licorice could relieve the oxidative stress injury induced by metronidazole (MTZ) and could restrain the production of excessive reactive oxygen species (ROS) induced by 2,2′-azobis (2-methylpropionamidine) dihydrochloride (AAPH) in a zebrafish model. It was further found that roasted licorice could exert its oxidative activity by upregulating the expression of key genes such as heme oxygenase 1 (HO-1), NAD(P)H quinone dehydrogenase 1 (NQO1), glutamate–cysteine ligase modifier subunit (GCLM), and glutamate–cysteine ligase catalytic subunit (GCLC) in the nuclear factor erythroid 2-related factor 2 (NRF2) signaling pathway both in vivo and in vitro. Furthermore, consistent results were obtained showing that rat serum containing roasted licorice was estimated to reduce cell apoptosis induced by H₂O₂. Then, the UHPLC-Q-Exactive Orbitrap MS analysis results elucidated the chemical composition of rat plasma containing roasted licorice extracts, including ten prototype chemical components and five metabolic components. Among them, six compounds were found to have binding activity with Kelch-like ECH-associated protein 1 (KEAP1), which plays a crucial role in the transcriptional activity of NRF2, using a molecular docking simulation. The results also showed that liquiritigenin had the strongest binding ability with KEAP1. Immunofluorescence further confirmed that liquiritigenin could induce the nuclear translocation of NRF2. In summary, this study provides a better understanding of the antioxidant effect and mechanisms of roasted licorice, and lays a theoretical foundation for the development of a potential antioxidant for use in clinical practice. Full article

Alzheimer’s disease (AD), a main cause of dementia, is the most common neurodegenerative disease that is related to the abnormal accumulation of amyloid β (Aβ) proteins. Yi-Gan-San (YGS), a traditional herbal medicine, has been used for the management of neurodegenerative disorders and for the treatment of neurosis, insomnia and dementia. The aim of this study was to examine antioxidant capacity and cytotoxicity of YGS treatment by using 2,2-Diphenyl-1-picrylhydrazyl (DPPH) and 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays in vitro. We explored neuroprotective effects of YGS treatment in alleviating Aβ neurotoxicity of Drosophila melanogaster in vivo by comparing survival rate, climbing index, and Aβ expressions through retinal green fluorescent protein (GFP) expression, highly sensitive immunomagnetic reduction (IMR) and Western blotting assays. In the in vitro study, our results showed that scavenging activities of free radical and SH-SY5Y nerve cell viability were increased significantly (p < 0.01–0.05). In the in vivo study, Aβ₄₂-expressing flies (Aβ₄₂-GFP flies) and their WT flies (mCD8-GFP flies) were used as an animal model to examine the neurotherapeutic effects of YGS treatment. Our results showed that, in comparison with those Aβ₄₂ flies under sham treatments, Aβ₄₂ flies under YGS treatments showed a greater survival rate, better climbing speed, and lower Aβ₄₂ aggregation in Drosophila brain tissue (p < 0.01). Our findings suggest that YGS should have a beneficial alternative therapy for AD and dementia via alleviating Aβ neurotoxicity in the brain tissue. Full article

Sjögren’s syndrome (SS) is an inflammatory autoimmune disease primarily affecting the exocrine glands; it has a major impact on patients’ lives. The Chinese herbal formula SS-1 is composed of Gan Lu Yin, Sang Ju Yin, and Xuefu Zhuyu decoction, which exerts anti-inflammatory, immunomodulatory, and antifibrotic effects. Our previous study demonstrated that SS-1 alleviates clinical SS. This study aimed to evaluate the efficacy and mechanism of the Chinese herbal formula SS-1 for salivary gland protein-induced experimental Sjögren’s syndrome (ESS). These results showed that ESS treatment with the Chinese herbal formula SS-1 (1500 mg/kg) significantly alleviated the severity of ESS. We found that SS-1 substantially improved saliva flow rates in SS mice and ameliorated lymphocytic infiltrations in submandibular glands. In addition, salivary gland protein-induced SS in mice treated with SS-1 significantly lowered proinflammatory cytokines (including IFN-γ, IL-6, and IL-17A) in mouse salivary glands and decreased serum anti-M3R autoantibody levels. In addition, we found that CD4+ T cells isolated from SS-1-treated SS mice significantly reduced the percentages of IFN-γ-producing CD4+ T cells (Th1) and IL-17A-producing CD4+ T cells (Th17). Our data show that SS-1 alleviates ESS through anti-inflammatory and immunomodulatory effects, which provides new insight into the clinical treatment of SS. Full article

Computer-assisted de novo design of natural product mimetics offers a viable strategy to reduce synthetic efforts and obtain natural-product-inspired bioactive small molecules, but suffers from several limitations. Deep learning techniques can help address these shortcomings. We propose the generation of synthetic molecule structures that optimizes the binding affinity to a target. To achieve this, we leverage important advancements in deep learning. Our approach generalizes to systems beyond the source system and achieves the generation of complete structures that optimize the binding to a target unseen during training. Translating the input sub-systems into the latent space permits the ability to search for similar structures, and the sampling from the latent space for generation. Full article