A comprehensive review and comparison of existing computational methods for protein function prediction.

Protein function prediction is critical for understanding the cellular physiological and biochemical processes, and it opens up new possibilities for advancements in fields such as disease research and drug discovery. During the past decades, with the exponential growth of protein sequence data, many computational methods for predicting protein function have been proposed. Therefore, a systematic review and comparison of these methods are necessary. In this study, we divide these methods into four different categories, including sequence-based methods, 3D structure-based methods, PPI network-based methods and hybrid information-based methods. Furthermore, their advantages and disadvantages are discussed, and then their performance is comprehensively evaluated and compared. Finally, we discuss the challenges and opportunities present in this field.

Preclinical Therapeutic Efficacy of RAF/MEK/ERK and IGF1R/AKT/mTOR Inhibition in Neuroblastoma.

Activating mutations in the RAS/MAPK pathway are observed in relapsed neuroblastoma. Preclinical studies indicate that these tumors have an increased sensitivity to inhibitors of the RAS/MAPK pathway, such as MEK inhibitors. MEK inhibitors do not induce durable responses as single agents, indicating a need to identify synergistic combinations of targeted agents to provide therapeutic benefit. We previously showed preclinical therapeutic synergy between a MEK inhibitor, trametinib, and a monoclonal antibody specific for IGF1R, ganitumab in RAS-mutated rhabdomyosarcoma. Neuroblastoma cells, like rhabdomyosarcoma cells, are sensitive to the inhibition of the RAS/MAPK and IGF1R/AKT/mTOR pathways. We hypothesized that the combination of trametinib and ganitumab would be effective in RAS-mutated neuroblastoma. In this study, trametinib and ganitumab synergistically suppressed neuroblastoma cell proliferation and induced apoptosis in cell culture. We also observed a delay in tumor initiation and prolongation of survival in heterotopic and orthotopic xenograft models treated with trametinib and ganitumab. However, the growth of both primary and metastatic tumors was observed in animals receiving the combination of trametinib and ganitumab. Therefore, more preclinical work is necessary before testing this combination in patients with relapsed or refractory RAS-mutated neuroblastoma.

Chlorogenic Acid as a Potential Therapeutic Agent for Cholangiocarcinoma.

Chlorogenic acid (CGA) has demonstrated anti-tumor effects across various cancers, but its role in cholangiocarcinoma (CCA) remains unclear. Our study revealed CGA's potent anti-tumor effects on CCA, significantly suppressing cell proliferation, migration, colony formation, and invasion while inhibiting the epithelial-mesenchymal transition. CGA induced apoptosis, modulated cell cycle progression, and exhibited a stable binding affinity to AKR1B10 in CCA. AKR1B10 was highly expressed in RBE cells, and CGA treatment reduced AKR1B10 expression. Knocking out AKR1B10 inhibited the proliferation of RBE cells, whereas the overexpression of AKR1B10 promoted their proliferation. Additionally, CGA suppressed the proliferation of RBE cells with AKR1B10 overexpression. Mechanistically, AKR1B10 activated AKT, and CGA exerted its inhibitory effect by reducing AKR1B10 levels, thereby suppressing AKT activation. Furthermore, CGA facilitated the polarization of tumor-associated macrophages towards an anti-tumor phenotype and enhanced T-cell cytotoxicity. These findings underscore CGA's potential as a promising therapeutic agent for CCA treatment.

RDLR: A Robust Deep Learning-Based Image Registration Method for Pediatric Retinal Images.

Retinal diseases stand as a primary cause of childhood blindness. Analyzing the progression of these diseases requires close attention to lesion morphology and spatial information. Standard image registration methods fail to accurately reconstruct pediatric fundus images containing significant distortion and blurring. To address this challenge, we proposed a robust deep learning-based image registration method (RDLR). The method consisted of two modules: registration module (RM) and panoramic view module (PVM). RM effectively integrated global and local feature information and learned prior information related to the orientation of images. PVM was capable of reconstructing spatial information in panoramic images. Furthermore, as the registration model was trained on over 280,000 pediatric fundus images, we introduced a registration annotation automatic generation process coupled with a quality control module to ensure the reliability of training data. We compared the performance of RDLR to the other methods, including conventional registration pipeline (CRP), voxel morph (WM), generalizable image matcher (GIM), and self-supervised techniques (SS). RDLR achieved significantly higher registration accuracy (average Dice score of 0.948) than the other methods (ranging from 0.491 to 0.802). The resulting panoramic retinal maps reconstructed by RDLR also demonstrated substantially higher fidelity (average Dice score of 0.960) compared to the other methods (ranging from 0.720 to 0.783). Overall, the proposed method addressed key challenges in pediatric retinal imaging, providing an effective solution to enhance disease diagnosis. Our source code is available at https://github.com/wuwusky/RobustDeepLeraningRegistration .

BTLA deficiency promotes HSC activation and protects against hepatic ischemia-reperfusion injury.

BACKGROUND AND AIMS: Hepatic ischemia-reperfusion injury (IRI) is unavoidable even despite the development of more effective surgical approaches. During hepatic IRI, activated HSC (aHSC) are involved in liver injury and recovery. APPROACH AND RESULT: A proportion of aHSC increased significantly both in the mouse liver tissues with IRI and in the primary mouse HSCs and LX-2 cells during hypoxia-reoxygenation. "Loss-of-function" experiments revealed that depleting aHSC with gliotoxin exacerbated liver damage in IRI mice. Subsequently, we found that the transcription of mRNA and the expression of B and T lymphocyte attenuator (BTLA) protein were lower in aHSC compared with quiescent HSCs. Interestingly, overexpression or knockdown of BTLA resulted in opposite changes in the activation of specific markers for HSCs such as collagen type I alpha 1, α-smooth muscle actin, and Vimentin. Moreover, the upregulation of these markers was also observed in the liver tissues of global BLTA-deficient (BTLA-/-) mice and was higher after hepatic IRI. Compared with wild-type mice, aHSC were higher, and liver injury was lower in BTLA-/- mice following IRI. However, the depletion of aHSC reversed these effects. In addition, the depletion of aHSC significantly exacerbated liver damage in BTLA-/- mice with hepatic IRI. Furthermore, the TGF-ß1 signaling pathway was identified as a potential mechanism for BTLA to negatively regulate the activation of HSCs in vivo and in vitro. CONCLUSIONS: These novel findings revealed a critical role of BTLA. Particularly, the receptor inhibits HSC-activated signaling in acute IRI, implying that it is a potential immunotherapeutic target for decreasing the IRI risk.

A Lesion-Fusion Neural Network for Multi-View Diabetic Retinopathy Grading.

As the most common complication of diabetes, diabetic retinopathy (DR) is one of the main causes of irreversible blindness. Automatic DR grading plays a crucial role in early diagnosis and intervention, reducing the risk of vision loss in people with diabetes. In these years, various deep-learning approaches for DR grading have been proposed. Most previous DR grading models are trained using the dataset of single-field fundus images, but the entire retina cannot be fully visualized in a single field of view. There are also problems of scattered location and great differences in the appearance of lesions in fundus images. To address the limitations caused by incomplete fundus features, and the difficulty in obtaining lesion information. This work introduces a novel multi-view DR grading framework, which solves the problem of incomplete fundus features by jointly learning fundus images from multiple fields of view. Furthermore, the proposed model combines multi-view inputs such as fundus images and lesion snapshots. It utilizes heterogeneous convolution blocks (HCB) and scalable self-attention classes (SSAC), which enhance the ability of the model to obtain lesion information. The experimental results show that our proposed method performs better than the benchmark methods on the large-scale dataset.

Recent advances in functional nucleic acid decorated nanomaterials for cancer imaging and therapy.

Nanomaterials possess unusual physicochemical properties including unique optical, magnetic, electronic properties, and large surface-to-volume ratio. However, nanomaterials face some challenges when they were applied in the field of biomedicine. For example, some nanomaterials suffer from the limitations such as poor selectivity and biocompatibility, low stability, and solubility. To address the above-mentioned obstacles, functional nucleic acid has been widely served as a powerful and versatile ligand for modifying nanomaterials because of their unique characteristics, such as ease of modification, excellent biocompatibility, high stability, predictable intermolecular interaction and recognition ability. The functionally integrating functional nucleic acid with nanomaterials has produced various kinds of nanocomposites and recent advances in applications of functional nucleic acid decorated nanomaterials for cancer imaging and therapy were summarized in this review. Further, we offer an insight into the future challenges and perspectives of functional nucleic acid decorated nanomaterials.

Naked-Eye 3-Dimensional Vision Training for Myopia Control: A Randomized Clinical Trial.

Importance: Early onset of myopia increases the risk of high myopia, which can lead to irreversible retinal damage and even loss of central vision. Objective: To investigate the efficacy and safety of naked-eye 3-dimensional vision training (NVT) in preventing the progression of myopia in children. Design, Setting, and Participants: This randomized clinical trial was conducted in 3 hospitals from May 25, 2022, to February 24, 2023. Participants were children (aged 6-18 years) who had a diagnosis of myopia with a spherical equivalent refraction of -0.75 to -6.00 diopters (D). Intervention: Children in the intervention group received 20 minutes of NVT treatment every day, whereas children in the control group lived as usual without vision training. Main Outcome and Measure: The primary outcome was the change in axial length at 6 months. Spherical equivalent refraction (SER) was included as a secondary outcome. Results: Among 263 participants, 125 (47.5%) were male and 138 (52.5%) were female; the mean (SD) age was 10.3 (1.9) years (range, 6.1-15.6 years). A total of 227 patients (86.3%) completed the 6-month follow-up, including 102 in the intervention group and 125 in the control group. In the intervention group, the changes in axial length and SER at 6 months were 0.18 mm (95% CI, 0.16 to 0.20 mm) and -0.25 D (95% CI, -0.31 to -0.19 D), respectively. In the control group, the changes in axial length and SER at 6 months were 0.23 mm (95% CI, 0.21 to 0.25 mm) and -0.35 D (95% CI, -0.41 to -0.30 D), respectively. The differences in AL and SER between the 2 groups were significant (AL difference: -0.06 mm; 95% CI, -0.09 to -0.03; P < .001; SER difference: 0.10 D; 95% CI, 0.02 to 0.19; P = .02). No study-related adverse reactions were reported during follow-up. Conclusions and Relevance: NVT is a safe and promising means to control myopia progression in children with good adherence. Trial Registration: ClinicalTrials.gov Identifier: NCT05468775.

Choriocapillaris flow features in children with myopic anisometropia.

AIMS: To examine differences between the eyes in choriocapillaris perfusion and choroidal thickness in children with myopic anisometropia. METHODS: In this observational and prospective study, 46 children with myopic anisometropia were enrolled. Choriocapillaris perfusion parameters, including the percentage of flow voids, the total number of flow voids and the average flow void area were obtained by optical coherence tomography angiography (OCTA). The OCTA image was divided into a 1 mm-diameter central circle (C1) and a 2.5 mm-diameter annulus (without the inner central 1 mm circle, C1-2.5). Both C1 and C1-2.5 are centred on the foveola. The C1-2.5 was divided into nasal (N1-2.5), temporal (T1-2.5), inferior (I1-2.5) and superior (S1-2.5) areas. Differences in these parameters in different regions between eyes were analysed. RESULTS: There were no significant differences in the percentage of flow voids and the average flow void area between the fellow eyes. The total number of signal voids was significantly higher in the less myopic eyes in C1-2.5 (p=0.032), S1-2.5 (p=0.008) and N1-2.5 (p=0.019). Changes in spherical equivalent refraction and axial length were both correlated with the changes in the total number of flow voids in N1-2.5 (R=-0.431, p=0.03; R=-0.297, p=0.047). CONCLUSIONS: The choroid in the macular region becomes thinner and the total number of flow voids in the nasal macular region decreased with the amplitude of myopia. This suggests that a decrease in total number of flow voids may indicate an early change in myopia.

A universal ANN-to-SNN framework for achieving high accuracy and low latency deep Spiking Neural Networks.

Spiking Neural Networks (SNNs) have become one of the most prominent next-generation computational models owing to their biological plausibility, low power consumption, and the potential for neuromorphic hardware implementation. Among the various methods for obtaining available SNNs, converting Artificial Neural Networks (ANNs) into SNNs is the most cost-effective approach. The early challenges in ANN-to-SNN conversion work revolved around the susceptibility of converted SNNs to conversion errors. Some recent endeavors have attempted to mitigate these conversion errors by altering the original ANNs. Despite their ability to enhance the accuracy of SNNs, these methods lack generality and cannot be directly applied to convert the majority of existing ANNs. In this paper, we present a framework named DNISNM for converting ANN to SNN, with the aim of addressing conversion errors arising from differences in the discreteness and asynchrony of network transmission between ANN and SNN. The DNISNM consists of two mechanisms, Data-based Neuronal Initialization (DNI) and Signed Neuron with Memory (SNM), designed to respectively address errors stemming from discreteness and asynchrony disparities. This framework requires no additional modifications to the original ANN and can result in SNNs with improved accuracy performance, simultaneously ensuring universality, high precision, and low inference latency. We verify it experimentally on challenging object recognition datasets, including CIFAR10, CIFAR100, and ImageNet-1k. Experimental results show that the SNN converted by our framework has very high accuracy even at extremely low latency.

Microstructure and Friction Properties of AlCrTiVNbx High-Entropy Alloys via Annealing Manufactured by Vacuum Arc Melting.

To enhance the friction and wear properties of alloys, AlCrTiVNbx high-entropy alloys (HEAs) with various Nb contents were prepared using the arc melting technique and then annealed at 1000 °C for 2 h. The microstructure and hardness changes in the AlCrTiVNbx (x = 0.3, 0.4, and 0.5) HEAs after casting and annealing were studied via scanning electron microscopy, X-ray diffractometry, optical microscopy and the Vickers hardness test. The MFT-EC400 ball disc reciprocating friction and wear tester was used to investigate the wear resistance of the HEAs before and after annealing. The results show that the annealed AlCrTiVNbx HEAs changed from a single-phase structure to a multi-phase structure, and the content of the face-center cubic (FCC) phase and hexagonal close-packed (HCP) phase further increases with the increase in Nb content. The hardness value of the annealed HEAs is greatly enhanced compared with the casting state, and the hardness of the Nb0.5 HEA is increased from 543 HV to 725 HV after annealing. The wear resistance of the alloys after the annealing treatment is also greatly improved, among which Nb0.5 has the best wear resistance. The average friction coefficient of Nb0.5 is 0.154 and the wear rate is 2.117 × 10-5 mm3/(N·m). We believe that the precipitation strengthening after the annealing treatment and the lubrication effect of the FCC phase are the reasons for the significant improvement in wear resistance. The morphology of the samples indicates that the wear mechanism of the alloy includes adhesive wear, abrasive wear and a certain degree of oxidation wear.

AMHGCN: Adaptive multi-level hypergraph convolution network for human motion prediction.

Human motion prediction is the key technology for many real-life applications, e.g., self-driving and human-robot interaction. The recent approaches adopt the unrestricted full-connection graph representation to capture the relationships inside the human skeleton. However, there are two issues to be solved: (i) these unrestricted full-connection graph representation methods neglect the inherent dependencies across the joints of the human body; (ii) these methods represent human motions using the features extracted from a single level and thus can neither fully exploit the various connection relationships among the human body nor guarantee the human motion prediction results to be reasonable. To tackle the above issues, we propose an adaptive multi-level hypergraph convolution network (AMHGCN), which uses the adaptive multi-level hypergraph representation to capture various dependencies among the human body. Our method has four different levels of hypergraph representations, including (i) the joint-level hypergraph representation to capture inherent kinetic dependencies in the human body, (ii) the part-level hypergraph representation to exploit the kinetic characteristics at a higher level (in comparison to the joint-level) by viewing some part of the human body as an entirety, (iii) the component-level hypergraph representation to model the semantic information, and (iv) the global-level hypergraph representation to extract long-distance dependencies in the human body. In addition, to take full advantage of the knowledge carried in the training data, we propose a reverse loss (i.e., adopting the future human poses to predict the historical poses reversely) to realize data augmentation. Extensive experiments show that our proposed AMHGCN can achieve state-of-the-art performance on three benchmarks, i.e., Human3.6M, CMU-Mocap, and 3DPW.

Regulation of Electronic Structures of the Urchin-Like NiCoP/CoP Nanocatalysts for Fast Hydrogen Evolution.

The exploration of stable, efficient, and low-cost catalysts toward ammonia borane hydrolysis is of vital significance for the practical implementation of this hydrogen production technology. Integrating interface engineering and nano-architecture engineering is a favorable strategy to elevate catalytic performance, as it can modify the electronic structure and provide sufficient active sites simultaneously. In this work, urchin-like NiCoP/CoP heterostructures are prepared via a three-step hydrothermal-oxidation-phosphorization synthesis route. It is demonstrated that the original Ni/Co molar ratio and the amount of phosphorus are crucial for adjusting the morphology, enhancing the exposed surface area, facilitating charge transfer, and modulating the adsorption and activation of H2O molecules. Consequently, the optimal Ni1Co2P heterostructure displays remarkable catalytic properties in the hydrolysis of ammonia borane with a turnover frequency (TOF) value of 30.3 molH2 ⋅ min-1 ⋅ molmetal -1, a low apparent activation energy of 25.89 kJ ⋅ mol-1, and good stability. Furthermore, by combining infrared spectroscopy and isotope kinetics experiments, a possible mechanism for the hydrolysis of ammonia borane was proposed.

Silencing of KIF2C enhances the sensitivity of hepatocellular carcinoma cells to cisplatin through regulating the PI3K/AKT/MAPK signaling pathway.

In the treatment of unresectable advanced hepatocellular carcinoma (HCC), cisplatin is administered transhepatic arterially for local treatment, but the clinical application of cisplatin drugs is frequently hindered by the emergence of drug resistance. Kinesin family member 2C( KIF2C ) has been shown as oncogene in a variety of tumors. Nevertheless, its effect on cisplatin sensitivity has yet to be ascertained. Herein, we aim to investigate the impact of the KIF2C gene on cisplatin sensitivity within HCC and the plausible underlying molecular mechanism. We examined the expression level of the KIF2C gene in HCC cells by real-time quantitative reverse transcription PCR and Western blot analysis, and analyzed bioinformatically by The Gene Expression Omnibus database and The Cancer Genome Atlas database. The KIF2C gene was silenced using the small interfering RNA technology, and its effect on cisplatin drug sensitivity in HCC cells was evaluated by flow cytometry, cell proliferation, cell migration, and invasion assays. Our results indicated that KIF2C was highly expressed in HCC cells. KIF2C silencing inhibits HCC cell proliferation, migration and invasion, promotes apoptosis, and keeps the cell cycle in G2 phase. In addition, KIF2C silencing enhanced the sensitivity of HCC cells to cisplatin. KIF2C silencing down-regulates the expression levels of phosphatidylinositol 3-kinase (PI3K), protein kinase B (AKT) and mitogen-activated protein kinase 3 (MAPK3) proteins. In conclusion, KIF2C silencing amplifies the sensitivity of HCC cells to cisplatin by regulating the PI3K/AKT/MAPK signaling pathway. Consequently, targeting KIF2C shows great application potential as a strategy for enhancing the effectiveness of HCC treatment.

Biomass carbon and Ti2C3MXene quantum dots as ratiometric fluorescent probes for sensitive detecting malachite green in fish sample.

A visual detection method for malachite green (MG) in food was established based on 'double-response-OFF' ratiometric fluorescent paper-based sensor. Biomass carbon quantum dots (BCQDs) using broad bean shell, and Ti3C2MXene quantum (MQDs) dots modified by ethylenediamine were synthesized by solvothermal method. The MG and two kinds of quantum dots could undergo static quenching, and the fluorescence color of two kinds of quantum dots gradually changed from red to blue, eventually the fluorescence was quenched, and the pattern had a two-stage linear relationship using fluorescent spectrofluorometer in the range of 0.1-140.0µM and the detection limit of 0.07µM. On this basis, a BCQDs/MQDs ratiometric fluorescence paper-based sensor was constructed and applied to fish sample. Through mobile phone software-Color recognizer, RGB values of fluorescent paper-based sensor at various concentrations of MG were extracted. The results showed that MG concentration was linearly correlated withR' value of RGB in the range of 20.0-140.0µM with 16.5µM detection limit. The method had been applied to the determination of canned fish and fresh basa fish samples, and the recovery rates were 97.33%-108.93% and 96.04%-117.97%, respectively. It proved that the ratiometric fluorescent paper-based sensor could be used for the rapid visual quantitative detecting MG in real samples.