Toward an Explainable Large Language Model for the Automatic Identification of the Drug-Induced Liver Injury Literature.

Drug-induced liver injury (DILI) stands as a significant concern in drug safety, representing the primary cause of acute liver failure. Identifying the scientific literature related to DILI is crucial for monitoring, investigating, and conducting meta-analyses of drug safety issues. Given the intricate and often obscure nature of drug interactions, simple keyword searching can be insufficient for the exhaustive retrieval of the DILI-relevant literature. Manual curation of DILI-related publications demands pharmaceutical expertise and is susceptible to errors, severely limiting throughput. Despite numerous efforts utilizing cutting-edge natural language processing and deep learning techniques to automatically identify the DILI-related literature, their performance remains suboptimal for real-world applications in clinical research and regulatory contexts. In the past year, large language models (LLMs) such as ChatGPT and its open-source counterpart LLaMA have achieved groundbreaking progress in natural language understanding and question answering, paving the way for the automated, high-throughput identification of the DILI-related literature and subsequent analysis. Leveraging a large-scale public dataset comprising 14 203 training publications from the CAMDA 2022 literature AI challenge, we have developed what we believe to be the first LLM specialized in DILI analysis based on LLaMA-2. In comparison with other smaller language models such as BERT, GPT, and their variants, LLaMA-2 exhibits an enhanced out-of-fold accuracy of 97.19% and area under the ROC curve of 0.9947 using 3-fold cross-validation on the training set. Despite LLMs' initial design for dialogue systems, our study illustrates their successful adaptation into accurate classifiers for automated identification of the DILI-related literature from vast collections of documents. This work is a step toward unleashing the potential of LLMs in the context of regulatory science and facilitating the regulatory review process.

Dissipation and residues of imidacloprid in amaranth under greenhouse and open field cultivations.

Despite the extensive exposure to imidacloprid residues in food plants, there has been little research on imidacloprid residues in amaranth. The dissipation trend and residue behavior of imidacloprid were evaluated to provide guidelines for imidacloprid application on amaranth under open field and greenhouse. The dissipation rate of imidacloprid in amaranth conformed to the first-order kinetic equation, and the half-lives of imidacloprid in amaranth ranged from 0.29 days in open field to 1.29 days in the greenhouse. After 7 and 14 days from the application of imidacloprid (pesticide dosage, 45 or 67.5 g a.i./ha), the amaranth under the open field and greenhouse growth could be consumed safely with average residues of 0.19 and 0.38 mg/kg, respectively. This result demonstrated that the cultivation has the dominant influence on imidacloprid residue, and the residue of imidacloprid in amaranth planting on open field was much lower than that in the greenhouse, indicating a significant difference in the pesticide residues between the two cultivations with a p-value less than 0.05.

The impact of long-term antihypertensive treatment on wound healing after major non-cardiac surgery in patients with cardiovascular diseases: A meta-analysis.

Hypertension is a prevalent condition that poses significant challenges in the perioperative management of patients undergoing major non-cardiac surgery, particularly concerning wound healing and scar formation. This meta-analysis assesses the impact of long-term antihypertensive treatment on postoperative wound healing, examining data from seven studies involving patients who received such treatments compared to untreated controls. Our findings reveal that long-term antihypertensive therapy is associated with significantly improved wound healing outcomes, as indicated by lower REEDA scores (I2 = 96%, SMD = -25.71, 95% CI: [-33.71, -17.70], p < 0.01) 1 week post-surgery and reduced scar formation, demonstrated by lower Manchester Scar Scale scores (I2 = 93%, SMD = -37.29, 95% CI: [-44.93, -29.64], p < 0.01) 2 months post-surgery. These results underscore the potential benefits of antihypertensive treatment in enhancing surgical recovery and offer insights into optimising perioperative care for hypertensive patients.

TGF-ß signaling pathway in spinal cord injury: Mechanisms and therapeutic potential.

Spinal cord injury (SCI) is a highly disabling central nervous system injury with a complex pathological process, resulting in severe sensory and motor dysfunction. The current treatment modalities only alleviate its symptoms and cannot effectively intervene or treat its pathological process. Many studies have reported that the transforming growth factor (TGF)-ß signaling pathway plays an important role in neuronal differentiation, growth, survival, and axonal regeneration after central nervous system injury. Furthermore, the TGF-ß signaling pathway has a vital regulatory role in SCI pathophysiology and neural regeneration. Following SCI, regulation of the TGF-ß signaling pathway can suppress inflammation, reduce apoptosis, prevent glial scar formation, and promote neural regeneration. Due to its role in SCI, the TGF-ß signaling pathway could be a potential therapeutic target. This article reported the pathophysiology of SCI, the characteristics of the TGF-ß signaling pathway, the role of the TGF-ß signaling pathway in SCI, and the latest evidence for targeting the TGF-ß signaling pathway for treating SCI. In addition, the limitations and difficulties in TGF-ß signaling pathway research in SCI are discussed, and solutions are provided to address these potential challenges. We hope this will provide a reference for the TGF-ß signaling pathway and SCI research, offering a theoretical basis for targeted therapy of SCI.

SIRT1 is a regulator of autophagy: Implications for the progression and treatment of myocardial ischemia-reperfusion.

SIRT1 is a highly conserved nicotinamide adenine dinucleotide (NAD+)-dependent histone deacetylase. It is involved in the regulation of various pathophysiological processes, including cell proliferation, survival, differentiation, autophagy, and oxidative stress. Therapeutic activation of SIRT1 protects the heart and cardiomyocytes from pathology-related stress, particularly myocardial ischemia/reperfusion (I/R). Autophagy is an important metabolic pathway for cell survival during energy or nutrient deficiency, hypoxia, or oxidative stress. Autophagy is a double-edged sword in myocardial I/R injury. The activation of autophagy during the ischemic phase removes excess metabolic waste and helps ensure cardiomyocyte survival, whereas excessive autophagy during reperfusion depletes the cellular components and leads to autophagic cell death. Increasing research on I/R injury has indicated that SIRT1 is involved in the process of autophagy and regulates myocardial I/R. SIRT1 regulates autophagy through various pathways, such as the deacetylation of FOXOs, ATGs, and LC3. Recent studies have confirmed that SIRT1-mediated autophagy plays different roles at different stages of myocardial I/R injury. By targeting the mechanism of SIRT1-mediated autophagy at different stages of I/R injury, new small-molecule drugs, miRNA activators, or blockers can be developed. For example, resveratrol, sevoflurane, quercetin, and melatonin in the ischemic stage, coptisine, curcumin, berberine, and some miRNAs during reperfusion, were involved in regulating the SIRT1-autophagy axis, exerting a cardioprotective effect. Here, we summarize the possible mechanisms of autophagy regulation by SIRT1 in myocardial I/R injury and the related molecular drug applications to identify strategies for treating myocardial I/R injury.

A prediction model for blood-brain barrier penetrating peptides based on masked peptide transformers with dynamic routing.

Blood-brain barrier penetrating peptides (BBBPs) are short peptide sequences that possess the ability to traverse the selective blood-brain interface, making them valuable drug candidates or carriers for various payloads. However, the in vivo or in vitro validation of BBBPs is resource-intensive and time-consuming, driving the need for accurate in silico prediction methods. Unfortunately, the scarcity of experimentally validated BBBPs hinders the efficacy of current machine-learning approaches in generating reliable predictions. In this paper, we present DeepB3P3, a novel framework for BBBPs prediction. Our contribution encompasses four key aspects. Firstly, we propose a novel deep learning model consisting of a transformer encoder layer, a convolutional network backbone, and a capsule network classification head. This integrated architecture effectively learns representative features from peptide sequences. Secondly, we introduce masked peptides as a powerful data augmentation technique to compensate for small training set sizes in BBBP prediction. Thirdly, we develop a novel threshold-tuning method to handle imbalanced data by approximating the optimal decision threshold using the training set. Lastly, DeepB3P3 provides an accurate estimation of the uncertainty level associated with each prediction. Through extensive experiments, we demonstrate that DeepB3P3 achieves state-of-the-art accuracy of up to 98.31% on a benchmarking dataset, solidifying its potential as a promising computational tool for the prediction and discovery of BBBPs.

Roles of pyroptosis in intervertebral disc degeneration.

Intervertebral disc degeneration (IDD), the key pathological process in low back pain, is characterized by chronic inflammation and progressive cell death. Pyroptosis is a type of pro-inflammatory programmed necrosis mediated by inflammasomes that is dependent on the gasdermin family of proteins. An in-depth study of the pathological mechanisms of IDD has revealed that pyroptosis plays an important role in its occurrence and development. The molecular characteristics and activation signaling mechanisms of pyroptosis are reviewed in this paper. Moreover, the specific roles of pyroptosis in IDD pathology are outlined and various targeted drugs for its treatment are highlighted.

The Different Welding Layers and Heat Source Energy on Residual Stresses in Swing Arc Narrow Gap MAG Welding.

In this paper, in order to reduce the time cost of prediction experiments in industry, a new narrow gap oscillation calculation method is developed in ABAQUS thermomechanical coupling analysis to study the distribution trend of residual weld stresses in comparison with conventional multi-layer welding processes. The blind hole detection technique and thermocouple measurement method verify the reliability of the prediction experiment. The results show that the experimental and simulation results have a high degree of agreement. In the prediction experiments, the calculation time of the high-energy single-layer welding experiments is 1/4 of the traditional multi-layer welding. Two welding processes of longitudinal residual stress and transverse residual stress distribution trends are the same. The high-energy single-layer welding experiment stress distribution range and transverse residual stress peak are smaller, but the longitudinal residual stress peak is slightly higher, which can be effectively reduced by increasing the preheating temperature of the welded parts. This implies that in the specific case of increasing the initial temperature of the workpiece, the use of high-energy single-layer welding instead of multi-layer welding to study the residual stress distribution trend not only optimizes the weld quality but also reduces the time cost to a large extent.

The role of long non-coding RNA HCG18 in cancer

The incidence of cancer is increasing worldwide and is becoming the most common cause of death. Identifying new biomarkers for cancer diagnosis and prognosis is important for developing cancer treatment strategies and reducing mortality. Long non-coding RNAs (lncRNAs) are non-coding, single-stranded RNAs that play an important role as oncogenes or tumor suppressors in the occurrence and development of human tumors. Abnormal expression of human leukocyte antigen complex group 18 (HCG18) is observed in many types of cancer, and its imbalance is closely related to cancer progression. HCG18 regulates cell proliferation, invasion, metastasis, and anti-apoptosis through a variety of mechanisms. Therefore, HCG18 is a potential tumor biomarker and therapeutic target. However, the therapeutic significance of HCG18 has not been well studied, and future research may develop new intervention strategies to combat cancer. In this study, we reviewed the biological function, mechanism, and potential clinical significance of HCG18 in various cancers to provide a reference for future research (AU)

The role of long non-coding RNA HCG18 in cancer.

The incidence of cancer is increasing worldwide and is becoming the most common cause of death. Identifying new biomarkers for cancer diagnosis and prognosis is important for developing cancer treatment strategies and reducing mortality. Long non-coding RNAs (lncRNAs) are non-coding, single-stranded RNAs that play an important role as oncogenes or tumor suppressors in the occurrence and development of human tumors. Abnormal expression of human leukocyte antigen complex group 18 (HCG18) is observed in many types of cancer, and its imbalance is closely related to cancer progression. HCG18 regulates cell proliferation, invasion, metastasis, and anti-apoptosis through a variety of mechanisms. Therefore, HCG18 is a potential tumor biomarker and therapeutic target. However, the therapeutic significance of HCG18 has not been well studied, and future research may develop new intervention strategies to combat cancer. In this study, we reviewed the biological function, mechanism, and potential clinical significance of HCG18 in various cancers to provide a reference for future research.

Few-Shot Learning Geometric Ensemble for Multi-label Classification of Chest X-Rays.

This paper aims to identify uncommon cardiothoracic diseases and patterns on chest X-ray images. Training a machine learning model to classify rare diseases with multi-label indications is challenging without sufficient labeled training samples. Our model leverages the information from common diseases and adapts to perform on less common mentions. We propose to use multi-label few-shot learning (FSL) schemes including neighborhood component analysis loss, generating additional samples using distribution calibration and fine-tuning based on multi-label classification loss. We utilize the fact that the widely adopted nearest neighbor-based FSL schemes like ProtoNet are Voronoi diagrams in feature space. In our method, the Voronoi diagrams in the features space generated from multi-label schemes are combined into our geometric DeepVoro Multi-label ensemble. The improved performance in multi-label few-shot classification using the multi-label ensemble is demonstrated in our experiments (The code is publicly available at https://github.com/Saurabh7/Few-shot-learning-multilabel-cxray).

The role of Sirt3 in the changes of skeletal muscle mitophagy induced by hypoxic training.

We aimed to explore the role of Sirt3 in the regulation of skeletal muscle mitophagy with hypoxic training. C57BL/6J mice were randomly divided into four groups: C group (control), HT group (mice performed a hypoxic training of living in an environment with an oxygen concentration of 13.8% and treadmill exercise under normoxia for 6 weeks), T group (mice were subjected to an intraperitoneal (i.p.) injection of the Sirt3 inhibitor 3-(1H-1,2,3-triazol-4-yl) pyridine (3-TYP) 50 mg/kg three times per week for 6 weeks) and THT group (the hypoxic training of HT group with i.p. injection of 3-TYP in T group). The results showed that 6 weeks of hypoxic training could improve ATP synthesis in skeletal muscle. After the combined intervention of 3-TYP injection and hypoxic training, Sirt3, FOXO3a, and SOD2 protein contents were still lower than those in hypoxic training group. Hypoxic training cannot improve the negative effect of Sirt3 inhibition on muscle PINK1/Parkin signal. This study demonstrated that Sirt3 plays a key role in mediating skeletal muscle mitophagy by hypoxic training. The results of our study also provided the first evidence that mitophagy caused by hypoxic training might be transduced through the Sirt3-FOXO3a signaling pathway.

Hypoxic Training Ameliorates Skeletal Muscle Microcirculation Vascular Function in a Sirt3-Dependent Manner.

Hypoxic training improves the microcirculation function of human skeletal muscle, but its mechanism is still unclear. Silent information regulator 2 homolog 3 (Sirt3) can improve mitochondrial function and oxidative status. We aimed to examine the role of Sirt3 in the process of hypoxic training, which affects skeletal muscle microcirculation. C57BL/6 mice were assigned to control (C), hypoxic training (HT), Sirt3 inhibitor 3-(1H-1,2,3-triazol-4-yl) pyridine (3-TYP), and 3-TYP + hypoxic training (3-TYP + HT) groups (n = 6/group). Sirt3 inhibition was induced by intraperitoneal injection of Sirt3 inhibitor 3-TYP. After 6 weeks of intervention, microcirculatory capillary formation and vasomotor capacity were evaluated using immunofluorescence, Western blot, biochemical tests, and transmission electron microscopy (TEM). Laser Doppler flowmetry was used to evaluate skeletal muscle microcirculation blood flow characteristics. Six weeks of hypoxic training enhanced skeletal muscle microcirculation function and increased microcirculatory vasodilation capacity and capillary formation. After the pharmacological inhibition of Sirt3, the reserve capacity of skeletal muscle microcirculation was reduced to varying degrees. After the inhibition of Sirt3, mice completed the same hypoxic training, and we failed to observe the microcirculation function adaptation like that observed in hypoxic training alone. The microcirculation vasodilation and the capillaries number did not improve. Hypoxic training improved skeletal muscle microcirculation vasodilation capacity and increased skeletal muscle microcirculation capillary density. Sirt3 is involved in the adaptation of skeletal muscle microcirculation induced by hypoxic training.

Long non­coding RNA PART1: dual role in cancer.

Increasing evidence has shown that long non-coding RNAs (lncRNAs), which are non-coding endogenous single-stranded RNAs, play an essential role in various physiological and pathological processes through transcriptional interference, post-transcriptional regulation, and epigenetic modification. Moreover, lncRNAs, as oncogenes or tumor suppressor genes, play an important role in the occurrence and development of human cancers. Prostate androgen-regulated transcript 1 (PART1) was initially identified as a carcinogenic lncRNA in prostate adenomas. The upregulated expression of PART1 plays a tumor-promoting role in liver, prostate, lung cancers, and other tumors. In contrast, the expression of PART1 is downregulated in esophageal squamous cell carcinoma, glioma, and other tumors, which may inhibit the tumor. PART1 plays a dual role in cancer and regulates cell proliferation, apoptosis, invasion, and metastasis through a variety of potential mechanisms. These findings suggest that PART1 is a promising tumor biomarker and therapeutic target. This article reviews the biological functions, related mechanisms, and potential clinical significance of PART1 in a variety of human cancers.

[Exercise regulates lipid metabolism via lipophagy and its molecular mechanisms].

Lipophagy is a kind of selective autophagy, which can selectively identify and degrade lipid droplets and plays an important role in regulating cellular lipid metabolism and maintaining intracellular lipid homeostasis. Exercise can induce lipophagy and it is also an effective means of reducing body fat. In this review, we summarized the relationship between exercise and lipophagy in the liver, pancreas, adipose tissue, and the possible molecular mechanisms to provide a new clue for the prevention and treatment of fatty liver, obesity and other related metabolic diseases by exercise.

[Unfolded protein response signaling in mitochondria].

The mitochondrial unfolded protein response is an important component of the mitochondrial protein quality control program. It can effectively remove unfolded or misfolded proteins under stress, and maintain a stable and healthy mitochondrial pool. The mitochondrial unfolded protein response is coordinated by multiple signaling pathways. The classical ATF4/ATF5-CHOP pathway is induced by accumulation of unfolded or misfolded proteins in the mitochondrial matrix, which reduces stress toxicity by regulating molecular chaperones and proteases. Sirt3-FOXO3a-SOD2 pathway, located in the mitochondrial matrix, plays an important role in anti-oxidative damage. The ERα-NRF1-HTRA2 pathway mainly removes unfolded proteins in the mitochondrial membrane space and improves the quality control of mitochondrial proteins. These three signaling pathways work both independently and cooperatively to enhance mitochondrial capacity and maintain health under stress.

Biological Functions and Therapeutic Potential of Autophagy in Spinal Cord Injury.

Autophagy is an evolutionarily conserved lysosomal degradation pathway that maintains metabolism and homeostasis by eliminating protein aggregates and damaged organelles. Many studies have reported that autophagy plays an important role in spinal cord injury (SCI). However, the spatiotemporal patterns of autophagy activation after traumatic SCI are contradictory. Most studies show that the activation of autophagy and inhibition of apoptosis have neuroprotective effects on traumatic SCI. However, reports demonstrate that autophagy is strongly associated with distal neuronal death and the impaired functional recovery following traumatic SCI. This article introduces SCI pathophysiology, the physiology and mechanism of autophagy, and our current review on its role in traumatic SCI. We also discuss the interaction between autophagy and apoptosis and the therapeutic effect of activating or inhibiting autophagy in promoting functional recovery. Thus, we aim to provide a theoretical basis for the biological therapy of SCI.

Improved Peptide Retention Time Prediction in Liquid Chromatography through Deep Learning.

The accuracy of peptide retention time (RT) prediction model in liquid chromatography (LC) is still not sufficient for wider implementation in proteomics practice. Herein, we propose deep learning as an ideal tool to considerably improve this prediction. A new peptide RT prediction tool, DeepRT, was designed using a capsule network model, and the public data sets containing peptides separated by reverse-phase liquid chromatography were used to evaluate the DeepRT performance. Compared with other prevailing RT predictors, DeepRT attained overall improvement in the prediction of peptide RTs with an R2 of ∼0.994. Moreover, DeepRT was able to accommodate to the peptides that were separated by different types of LC, such as strong cation exchange (SCX) and hydrophilic interaction liquid chromatography (HILIC) and to reach the RT prediction with R2 values of ∼0.996 for SCX and ∼0.993 for HILIC, respectively. If a large peptide data set is available for one type of LC, DeepRT can be promoted to DeepRT(+) using transfer learning. Based on a large peptide data set gained from SWATH, DeepRT(+) further elevated the accuracy of RT prediction for peptides in a small data set and enabled a satisfactory prediction upon limited peptides approximating hundreds. Further, DeepRT automatically learns retention-related properties of amino acids under different separation mechanisms, which are well consistent with retention coefficients (Rc) of the amino acids. DeepRT was thus proven to be an improved RT predictor with high flexibility and efficiency. DeepRT is available at https://github.com/horsepurve/DeepRTplus .