Artificial intelligence-powered pharmacovigilance: A review of machine and deep learning in clinical text-based adverse drug event detection for benchmark datasets.

OBJECTIVE: The primary objective of this review is to investigate the effectiveness of machine learning and deep learning methodologies in the context of extracting adverse drug events (ADEs) from clinical benchmark datasets. We conduct an in-depth analysis, aiming to compare the merits and drawbacks of both machine learning and deep learning techniques, particularly within the framework of named-entity recognition (NER) and relation classification (RC) tasks related to ADE extraction. Additionally, our focus extends to the examination of specific features and their impact on the overall performance of these methodologies. In a broader perspective, our research extends to ADE extraction from various sources, including biomedical literature, social media data, and drug labels, removing the limitation to exclusively machine learning or deep learning methods. METHODS: We conducted an extensive literature review on PubMed using the query "(((machine learning [Medical Subject Headings (MeSH) Terms]) OR (deep learning [MeSH Terms])) AND (adverse drug event [MeSH Terms])) AND (extraction)", and supplemented this with a snowballing approach to review 275 references sourced from retrieved articles. RESULTS: In our analysis, we included twelve articles for review. For the NER task, deep learning models outperformed machine learning models. In the RC task, gradient Boosting, multilayer perceptron and random forest models excelled. The Bidirectional Encoder Representations from Transformers (BERT) model consistently achieved the best performance in the end-to-end task. Future efforts in the end-to-end task should prioritize improving NER accuracy, especially for 'ADE' and 'Reason'. CONCLUSION: These findings hold significant implications for advancing the field of ADE extraction and pharmacovigilance, ultimately contributing to improved drug safety monitoring and healthcare outcomes.

Effect of aspirin on blood pressure in hypertensive patients: a systematic review and meta-analysis.

INTRODUCTION: Aspirin is widely used for secondary prevention in patients with hypertension. However, previous studies mainly focused on the preventive effects of aspirin, and there has been a lack of reliable evidence on whether taking aspirin affects blood pressure This study aimed to investigate whether aspirin would affect the blood pressure in patients with hypertension. METHODS: PubMed, Cochrane database, Embase, Scopus and Medline databases were searched until September 2023. For continuous variables (e.g., blood pressure reduction), the mean difference (MD) was selected as the effect magnitude indices. We used the Cochrane Collaboration's Risk of Bias tool to assess the risk of bias. RESULT: A total of five studies were included, comprising 20,312 patients. We found that aspirin did not affect SBP (MD = -0.78, 95% CI: - 2.41, 0.84). A similar result was found for DBP (MD = -0.86, 95% CI: - 2.14, 0.42). CONCLUSION: This study showed no significant difference in blood pressure between the aspirin and control groups, suggesting that aspirin does not affect blood pressure.

Circ-ZNF236 mediates stem cells from apical papilla differentiation by regulating LGR4-induced autophagy.

AIM: Human stem cells from the apical papilla (SCAPs) are an appealing stem cell source for tissue regeneration engineering. Circular RNAs (circRNAs) are known to exert pivotal regulatory functions in various cell differentiation processes, including osteogenesis of mesenchymal stem cells. However, few studies have shown the potential mechanism of circRNAs in the odonto/osteogenic differentiation of SCAPs. Herein, we identified a novel circRNA, circ-ZNF236 (hsa_circ_0000857) and found that it was remarkably upregulated during the SCAPs committed differentiation. Thus, in this study, we showed the significance of circ-ZNF236 in the odonto/osteogenic differentiation of SCAPs and its underlying regulatory mechanisms. METHODOLOGY: The circular structure of circ-ZNF236 was identified via Sanger sequencing, amplification of convergent and divergent primers. The proliferation of SCAPs was detected by CCK-8, flow cytometry analysis and EdU incorporation assay. Western blotting, qRT-PCR, Alkaline phosphatase (ALP) and Alizarin red staining (ARS) were performed to explore the regulatory effect of circ-ZNF236/miR-218-5p/LGR4 axis in the odonto/osteogenic differentiation of SCAPs in vitro. Fluorescence in situ hybridization, as well as dual-luciferase reporting assays, revealed that circ-ZNF236 binds to miR-218-5p. Transmission electron microscopy (TEM) and mRFP-GFP-LC3 lentivirus were performed to detect the activation of autophagy. RESULTS: Circ-ZNF236 was identified as a highly stable circRNA with a covalent closed loop structure. Circ-ZNF236 had no detectable influence on cell proliferation but positively regulated SCAPs odonto/osteogenic differentiation. Furthermore, circ-ZNF236 was confirmed as a sponge of miR-218-5p in SCAPs, while miR-218-5p targets LGR4 mRNA at its 3'-UTR. Subsequent rescue experiments revealed that circ-ZNF236 regulates odonto/osteogenic differentiation by miR-218-5p/LGR4 in SCAPs. Importantly, circ-ZNF236 activated autophagy, and the activation of autophagy strengthened the committed differentiation capability of SCAPs. Subsequently, in vivo experiments showed that SCAPs overexpressing circ-ZNF236 promoted bone formation in a rat skull defect model. CONCLUSIONS: Circ-ZNF236 could activate autophagy through increasing LGR4 expression, thus positively regulating SCAPs odonto/osteogenic differentiation. Our findings suggested that circ-ZNF236 might represent a novel therapeutic target to prompt the odonto/osteogenic differentiation of SCAPs.

Association between sleep quality, depressive symptoms, and suicidal ideation in college students.

Suicide among college students is a challenging problem globally. Yet, the association between sleep quality, depressive symptoms, and suicidal ideation remains unclear. This study aims to understand if depressive symptoms mediate the relationship between sleep quality and suicide ideation and whether the interaction between depressive symptoms and sleep quality on suicidal ideation is additive. A total of 1182 college students were recruited, and sleep quality, depressive symptoms, and suicidal ideation were assessed using questionnaires. Univariate analysis, logistic regression analysis, linear regression models, and the Sobel test were performed. The results showed that, among college students, poor sleep quality was positively associated with suicidal ideation, and the association was mediated through depressive symptoms. Moreover, there was a significant additive interaction between poor sleep quality and depressive symptoms on suicidal ideation. These findings suggest that, in the process of preventing and treating suicidal ideation in college students with sleep disorders, we should focus on the evaluation and intervention of depressive symptoms and adopt multidisciplinary team interventions for college students with sleep disorders and depression.

PD-1 Suppresses the Osteogenic and Odontogenic Differentiation of Stem Cells from Dental Apical Papilla via Targeting SHP2/NF-κB Axis.

Stem cells from the apical papilla (SCAPs) are important for tooth root development and regeneration of root dentin. Here, we examined the expression of programmed cell death protein-1 (PD-1) in SCAPs and investigated the effects of PD-1 on odontogenic and osteogenic differentiation, as well as the relationship between PD-1 and SHP2/NF-κB signals. SCAPs were obtained and cultured in the related medium. The proliferation ability was evaluated by the cell counting kit 8 (CCK-8) and the 5-ethynyl-20-deoxyuridine (EdU) assay. Alkaline phosphatase (ALP) activity assay, ALP staining, Western blot, real-time quantitative reverse-transcription polymerase chain reaction (RT-qPCR), Alizarin Red S (ARS) staining, and immunofluorescence (IF) staining were performed to explore the osteo/odontogenic potential and the involvement of SHP2/NF-κB pathways. Besides, we transplanted SCAPs components into mouse calvaria defects to evaluate osteogenesis in vivo. We found that human SCAPs expressed PD-1 for the first time. PD-1 knockdown enhanced the osteo/odontogenic differentiation of SCAPs by suppressing the SHP2 pathway and activating the NF-κB pathway. Overexpression of PD-1 inhibited the osteogenesis and odontogenesis of SCAPs via activation of SHP2 signal and inhibition of the NF-κB pathway. PD-1 activated SHP2 signal to block NF-κB signal and then played a vital role in osteo/odontogenic differentiation of SCAPs.

SHED-derived exosomes attenuate trigeminal neuralgia after CCI of the infraorbital nerve in mice via the miR-24-3p/IL-1R1/p-p38 MAPK pathway.

BACKGROUND: Microglial activation in the spinal trigeminal nucleus (STN) plays a crucial role in the development of trigeminal neuralgia (TN). The involvement of adenosine monophosphate-activated protein kinase (AMPK) and N-methyl-D-aspartate receptor 1 (NMDAR1, NR1) in TN has been established. Initial evidence suggests that stem cells from human exfoliated deciduous teeth (SHED) have a potential therapeutic effect in attenuating TN. In this study, we propose that SHED-derived exosomes (SHED-Exos) may alleviate TN by inhibiting microglial activation. This study sought to assess the curative effect of SHED-Exos administrated through the tail vein on a unilateral infraorbital nerve chronic constriction injury (CCI-ION) model in mice to reveal the role of SHED-Exos in TN and further clarify the potential mechanism. RESULTS: Animals subjected to CCI-ION were administered SHED-Exos extracted by differential ultracentrifugation. SHED-Exos significantly alleviated TN in CCI mice (increasing the mechanical threshold and reducing p-NR1) and suppressed microglial activation (indicated by the levels of TNF-α, IL-1ß and IBA-1, as well as p-AMPK) in vivo and in vitro. Notably, SHED-Exos worked in a concentration dependent manner. Mechanistically, miR-24-3p-upregulated SHED-Exos exerted a more significant effect, while miR-24-3p-inhibited SHED-Exos had a weakened effect. Bioinformatics analysis and luciferase reporter assays were utilized for target gene prediction and verification between miR-24-3p and IL1R1. Moreover, miR-24-3p targeted the IL1R1/p-p38 MAPK pathway in microglia was increased in CCI mice, and participated in microglial activation in the STN. CONCLUSIONS: miR-24-3p-encapsulated SHED-Exos attenuated TN by suppressing microglial activation in the STN of CCI mice. Mechanistically, miR-24-3p blocked p-p38 MAPK signaling by targeting IL1R1. Theoretically, targeted delivery of miR-24-3p may offer a potential strategy for TN.

Hsa-miR-27b-5p suppresses the osteogenic and odontogenic differentiation of stem cells from human exfoliated deciduous teeth via targeting BMPR1A: An ex vivo study.

AIM: Recently, miR-27b-5p was shown to be abundantly expressed in extracellular vehicles (EVs) from the inflammatory microenvironment. This study determined the role of miR-27b-5p in regulating osteogenic and odontogenic differentiation of stem cells from human exfoliated deciduous teeth (SHEDs) and further examined the regulatory mechanism of bone morphogenetic protein receptor type-1A (BMPR1A). METHODOLOGY: Characteristics of SHEDs and SHEDs-EVs derived from SHEDs were evaluated respectively. The expression of miR-27b-5p in SHEDs and EVs was detected during osteo-induction. Mechanically, SHEDs were treated with miR-27b-5p mimics or an inhibitor, and the osteogenic/odontogenic differentiation and proliferation were assessed. Bioinformatic analysis and luciferase reporter were utilized for target gene prediction and verification. Finally, BMPR1A-overexpressed plasmids were transfected into SHEDs to investigate the participation of the BMPR1A/SMAD4 pathway. Data were analysed using Student's t-test, one-way analysis of variance and Chi-square test. RESULTS: MiR-27b-5p was expressed in both SHEDs and EVs and was significantly increased at the initial stage of differentiation and then decreased in a time-dependent manner (p < .01). Upregulation of miR-27b-5p significantly suppressed osteogenic/odontogenic differentiation of SHEDs and inhibited proliferation (p < .05), whereas inhibition of miR-27b-5p enhanced the differentiation (p < .05). Dual-luciferase reporter assay and pull-down assay confirmed the binding site between miR-27b-5p and BMPR1A (p < .05). The overexpression of BMPR1A rescued the effect of miR-27b-5p, while contributed to the decrease of pluripotency (p < .05). Additionally, miR-27b-5p maintained pluripotency in BMPR1A-overexpressed SHEDs (p < .05). CONCLUSIONS: MiR-27b-5p in SHEDs/EVs was inversely associated with differentiation and suppressed the osteogenic and odontogenic differentiation of SHEDs and maintained the pluripotency of SHEDs partly by shuttering BMPR1A-targeting BMP signalling. Theoretically, inhibition of miR-27b-5p represents a potential strategy to promote osteanagenesis and dentinogenesis. However, miR-27b-5p capsuled EVs might maintain cell pluripotency and self-renewal for non-cell-targeted therapy.

Temporal variability profiling of the default mode across epilepsy subtypes.

OBJECTIVE: Epilepsies are a group of neurological disorders sharing certain core features, but also demonstrate remarkable pathogenic and symptomatic heterogeneities. Various subtypes of epilepsy have been identified with abnormal shift in the brain default mode network (DMN). This study aims to evaluate the fine details of shared and distinct alterations in the DMN among epileptic subtypes. METHODS: We collected resting-state functional magnetic resonance imaging (MRI) data from a large epilepsy sample (n = 371) at a single center, including temporal lobe epilepsy (TLE), frontal lobe epilepsy (FLE), and genetic generalized epilepsy with generalized tonic-clonic seizures (GGE-GTCS), as well as healthy controls (HC, n = 150). We analyzed temporal dynamics profiling of the DMN, including edge-wise and node-wise temporal variabilities, and recurrent dynamic states of functional connectivity, to identify abnormalities common to epilepsies as well as those specific to each subtype. RESULTS: The analyses revealed that hypervariable edges within the specific DMN subsystem were shared by all subtypes (all PNBS  < .005), and deficits in node-wise temporal variability were prominent in TLE (all t(243) ≤ 2.51, PFDR  < .05) and FLE (all t(302) ≤ -2.65, PFDR  < .05) but relatively weak in GGE-GTCS. Moreover, dynamic states were generally less stable in patients than controls (all P's < .001). SIGNIFICANCE: Collectively, these findings demonstrated general DMN abnormalities common to different epilepsies as well as distinct dysfunctions to subtypes, and provided insights into understanding the relationship of pathophysiological mechanisms and brain connectivity.

Assessing Young Adults' ENDS Use via Ecological Momentary Assessment and a Smart Bluetooth Enabled ENDS Device.

INTRODUCTION: The assessment of electronic nicotine delivery systems (ENDS) use poses unique challenges that go beyond established assessment methods for tobacco cigarettes. Recent studies have proposed using ecological momentary assessment (EMA), a method to collect self-reported data on mobile devices, or data passively collected by "smart" Bluetooth enabled ENDS to assess use. The current study sought to compare ENDS use data using EMA and puff counts collected from a smart device. AIMS AND METHODS: We recruited 18 young adult ENDS users (age M = 23.33; 44.4% female) from the San Francisco Bay Area. For a total of 30 days, participants completed daily diaries by EMA and used a second-generation smart Bluetooth enabled ENDS that collected puff data. Repeated measures correlations, multilevel regressions, and paired t tests assessed concordance of EMA reports and ENDS data. A subset of four highly compliant participants were selected for sensitivity analyses. RESULTS: Among all 18 participants, completion of EMA daily diaries was high (77.4%). The ENDS device collected approximately twice as many puffs per day as participants reported. Compared with self-reported number of sessions and amount of e-liquid used, self-reported puff counts had the highest correlation with device-collected puff counts (rrm = 0.49; p < .001). Correlations between self-reported and device-collected puff counts improved among the subset of four highly compliant participants (rrm = 0.59; p < .001). CONCLUSIONS: Self-reports potentially underestimate use of ENDS. Puff counts appear to be the best self-reported measure to assess ENDS use compared with number of sessions or liquid volume. IMPLICATIONS: The comparison of EMA self-reports and passively collected ENDS device data can inform future efforts to assess ENDS use. Self-reported puff counts are preferable over number of sessions or amount of liquid used, but compared with objective usage data, self-reported puff counts may still underestimate actual use. ENDS use behavior is likely higher than users estimate and report. Future research on improved measures of ENDS use is needed.

CTP-CM enhances osteogenic differentiation of hPDLSCs via NF-κB pathway.

OBJECTIVE: The conditioned medium of calcined tooth powder (CTP-CM) is a type of biomimetic mineralized material and well contributing to bone healing and bone formation in vivo. However, little is known about the effect of CTP-CM on human periodontal ligament stem cells (hPDLSCs) as well as the underlying mechanisms. METHODS: ALP activity assay was conducted to select the concentration with the highest ALP level, which was used for the following experiments. Cell proliferation was measured by cell counting kit-8 assay and flow cytometry analysis. Expression levels of osteogenic markers in CTP-CM-induced hPDLSCs were evaluated with real-time quantitative reverse transcription polymerase chain reaction (qRT-PCR), immunofluorescence staining, and Western blot. Mineralization of CTP-CM-induced hPDLSCs was evaluated by alizarin red staining. Furthermore, the involvement of NF-κB pathway was examined by immunofluorescence staining and Western blot. RESULTS: 20 µg/ml was selected for the further experiments. Functional studies demonstrated that CTP-CM exerted almost no influence on the proliferation of hPDLSCs and CTP-CM increased the osteogenic differentiation of hPDLSCs. Mechanistically, CTP-CM leads to activation of NF-κB signaling pathway. When treated with BMS345541, the osteogenic differentiation of CTP-CM-treated hPDLSCs was significantly attenuated. CONCLUSION: CTP-CM can promote the osteogenic differentiation of hPDLSCs via activating NF-κB pathway.

Parathyroid hormone enhances the osteo/odontogenic differentiation of dental pulp stem cells via ERK and P38 MAPK pathways.

OBJECTIVE: Parathyroid hormone (PTH) is a main systemic mediator of calcium and phosphate homeostasis in the bone. Dental pulp stem cells (DPSCs) have been extensively studied in the regeneration of bone and tooth tissues. This paper aims to uncover the influences of PTH on the proliferative ability and osteo/odontogenic differentiation of DPSCs, as well as the underlying mechanisms. MATERIALS AND METHODS: The optimal concentration of PTH on DPSCs was determined by alkaline phosphatase (ALP) activity assay, ALP staining and western blot analysis. Proliferative ability and cell cycle distribution of DPSCs were analyzed by Cell counting kit-8, 5-ethynyl-20-deoxyuridine assay, and flow cytometry. Osteo/odontogenic capacity of DPSCs was evaluated and finally, the involvement of mitogen-activated protein kinase (MAPK) pathway was assessed. RESULTS: Purified DPSCs were obtained by enzymatic digestion, which presented a typical fibroblast-like morphology. 10-9 mol/L PTH was concerned as the optimal concentration for DPSCs induction. 10-9 mol/L PTH treatment did not change the proliferative rate of DPSCs (p > .05). Relative expressions of DSPP/DSPP, RUNX2/RUNX2, OSX/OSX, and ALP/ALP were upregulated in PTH-treated DPSCs relative to control group. Particularly, their mRNA/protein levels at Day 7 were markedly higher relative to those at Day 3 (p < .05 or p < .01). Mineralized nodules were formed after PTH induction, and calcium content increased by cetylpyridinium chloride quantitative analysis. Mechanistically, the protein levels of p-ERK and p-P38 significantly increased after PTH treatment, and the inhibitors targeting MAPK were identified that weakened the effects of PTH on the committed differentiation of DPSCs. CONCLUSIONS: PTH enhances the osteo/odontogenic differentiation capacity of DPSCs via ERK and P38 signaling pathways.

Impaired interactions among white-matter functional networks in antipsychotic-naive first-episode schizophrenia.

Schizophrenia has been conceptualized as a disorder arising from structurally pathological alterations to white-matter fibers in the brain. However, few studies have focused on white-matter functional changes in schizophrenia. Considering that converging evidence suggests that white-matter resting state functional MRI (rsfMRI) signals can effectively depict neuronal activity and psychopathological status, this study examined white-matter network-level interactions in antipsychotic-naive first-episode schizophrenia (FES) to facilitate the interpretation of the psychiatric pathological mechanisms in schizophrenia. We recruited 42 FES patients (FESs) and 38 healthy controls (HCs), all of whom underwent rsfMRI. We identified 11 white-matter functional networks, which could be further classified into deep, middle, and superficial layers of networks. We then examined network-level interactions among these 11 white-matter functional networks using coefficient Granger causality analysis. We employed group comparisons on the influences among 11 networks using network-based statistic. Excitatory influences from the middle superior corona radiate network to the superficial orbitofrontal and deep networks were disrupted in FESs compared with HCs. Additionally, an extra failure of suppression within superficial networks (including the frontoparietal network, temporofrontal network, and the orbitofrontal network) was observed in FESs. We additionally recruited an independent cohort (13 FESs and 13 HCs) from another center to examine the replicability of our findings across centers. Similar replication results further verified the white-matter functional network interaction model of schizophrenia. The novel findings of impaired interactions among white-matter functional networks in schizophrenia indicate that the pathophysiology of schizophrenia may also lie in white-matter functional abnormalities.

Potassium dihydrogen phosphate promotes the proliferation and differentiation of human periodontal ligament stem cells via nuclear factor kappa B pathway.

INTRODUCTION: Periodontal ligament stem cells (PDLSCs) are vital for the regeneration of periodontal tissues. Potassium dihydrogen phosphate (KH2PO4) has recently been applied as a component of the mineralization inducing medium (MM), which can be used to induce osteogenic differentiation of dental stem cells. However, whether KH2PO4 has effects on PDLSCs has not been studied. MATERIALS AND METHODS: PDLSCs were isolated by magnetic activated cell sorting and cultured. Alkaline phosphatase (ALP) activity and ALP protein expression of PDLSCs treated with different concentrations of KH2PO4 were examined to make sure the optimal concentration of KH2PO4 for the following experiments. The effects of KH2PO4 on the proliferation and differentiation of PDLSCs were investigated by flow cytometry, cell counting kit-8 assay, alizarin red staining, real-time RT-PCR, and Western blot. The involvement of nuclear factor kappa B (NF-κB) pathway in KH2PO4-treated PDLSCs was analyzed by Western blot and alizarin red staining. RESULTS: ALP activity assay and ALP protein expression examination revealed that 1.8 mmol/L KH2PO4 was the optimal concentration for the induction of hPDLSCs by KH2PO4. The proliferation and mineralization capacity of PDLSCs treated with KH2PO4 were enhanced as compared with the control group. PDLSCs treated with KH2PO4 showed an improved proliferation capacity in logarithmic growth phase at day 7. As PDLSCs were treated with KH2PO4, the expression of odonto/osteogenic markers (OCN/OCN, DSP/DSPP, OSX/OSX, RUNX2/RUNX2, and ALP/ALP) in cells were up-regulated at day 3 or 7. Moreover, the expression of IκBα in cytoplasm was down-regulated, along with an increased expression of p-P65 in cytoplasm and an up-regulated expression of P65 in nucleus. When treated with BMS345541 (the specific NF-κB inhibitor), the odonto/osteogenic differentiation of KH2PO4-treated PDLSCs was significantly attenuated. CONCLUSION: KH2PO4 can improve the proliferation and odonto/osteogenic differentiation capacity of PDLSCs via NF-κB pathway, and thus represents a potential target involved in the regeneration of periodontium for clinical treatments.

MiR-141-3p regulates proliferation and senescence of stem cells from apical papilla by targeting YAP.

Biological phenotypes of mesenchymal stem cells (MSCs) are regulated by a series of biochemical elements, including microRNAs, hormones and growth factors. Our previous study illustrated a significant role of miR-141-3p during the osteogenic differentiation of stem cells from apical papilla (SCAPs). Nevertheless, the functions of miR-141-3p in regulating the proliferative ability and senescence of SCAPs have not been determined. This study identified that overexpression of miR-141-3p inhibited the proliferative ability of SCAPs. Meanwhile, the senescence of SCAPs was ahead of time. Conversely, transfection of miR-141-3p inhibitor promoted the proliferative ability of SCAPs and delayed their senescence. Yes-associated protein (YAP) was predicted as the downstream target gene of miR-141-3p by online softwares (miRDB, miRTarBase, miRWalk, and TargetScan), and was further verified by dual-luciferase reporter gene assay. Additionally, knockdown of YAP inhibited the proliferation and accelerated the senescence of SCAPs. Collectively, these findings proposed a novel direction that miR-141-3p impeded proliferative ability and promoted senescence of SCAPs through post-transcriptionally downregulating YAP.

Sodium fluoride regulates the osteo/odontogenic differentiation of stem cells from apical papilla by modulating autophagy.

Fluoride (sodium fluoride) is thought to be essential in the development of tooth, and research shows that fluoride can modulate the differentiation of dental stem cells. However, the effects of fluoride on the committed differentiation of stem cells from apical papilla (SCAPs) and the underlying mechanisms remain unclear. Here, SCAPs were isolated from healthy extracted human third molars with immature roots and then were cultured with NaF conditioned media. Cell Counting Kit-8, EdU staining, and flow cytometry were performed to detected the proliferation activity. Alkaline phosphatase (ALP) activity, Alizarin Red staining, Western blot assay, and real-time reverse-transcription polymerase chain reaction were applied to assess the osteo/odontogenic differentiation NaF-treated SCAPs. Western blot assay and transmission electron microscope were used to evaluate the autophagy involved in the differentiation of SCAPs. ALP activity, ALP protein, and messenger RNA (mRNA) expression showed that 0.5 mM was the optimal concentration for the induction of SCAPs by NaF. 0.5 mM NaF-treated SCAPs induced more mineralized nodules as compared with untreated cells. Moreover, the osteo/odontogenic markers (RUNX2, OSX, DSP, and OCN) in mRNA levels were upregulated while the protein levels of these markers increased considerably in 0.5 mM NaF-treated SCAPs. Furthermore, the autophagy-related proteins (LC3, ATG5, and Beclin1) increased in NaF-treated SCAPs, and the osteo/odontogenic makers significantly decreased while silencing ATG5 to block autophagy. In all, sodium fluoride can regulate the osteo/odontogenic differentiation of SCAPs by modulating autophagy.

Modeling the effect of nanosecond laser conditioning on the femtosecond laser-induced damage of optical films.

The effect of nanosecond laser conditioning on the femtosecond laser-induced damage behaviors of Al2O3, HfO2, SiO2 single layers and Al2O3/SiO2 high reflectors (HR) are explored. During femtosecond laser damage test, negative effects on enhancing the femtosecond laser-induced damage threshold (LIDT) of optical films after the nanosecond laser conditioning is found, which is opposite to the LIDT improvement in the nanosecond range. To explain the mechanism after nanosecond laser conditioning, a theoretical model including multiphoton ionization (MPI), avalanche ionization (AI) and decays of electrons with one defect state is built to simulate the evolution of electron density in the conduction band. A permanent mid-gap defect state resulting from the process of laser conditioning is introduced in our model, which is found to contribute seed electrons to conduction band and hence accelerate the final breakdown. Both the experimental result and theoretical calculation agree very well with each other.

Ultrafast pre-breakdown dynamics in Al2O3SiO2 reflector by femtosecond UV laser spectroscopy.

Ultrafast carrier dynamics in Al2O3/SiO2 high reflectors has been investigated by UV femtosecond laser. It is identified by laser spectroscopy that, the carrier dynamics contributed from the front few layers of Al2O3 play a dominating role in the initial laser-induced damage of the UV reflector. Time-resolved reflection decrease after the UV excitation is observed, and conduction electrons is found to relaxed to a mid-gap defect state locating about one photon below the conduction band . To interpret the laser induced carrier dynamics further, a theoretical model including electrons relaxation to a mid-gap state is built, and agrees very well with the experimental results.. To the best of our knowledge, this is the first study on the pre-damage dynamics in UV high reflector induced by femtosecond UV laser.

Improving large language models for clinical named entity recognition via prompt engineering.

IMPORTANCE: The study highlights the potential of large language models, specifically GPT-3.5 and GPT-4, in processing complex clinical data and extracting meaningful information with minimal training data. By developing and refining prompt-based strategies, we can significantly enhance the models' performance, making them viable tools for clinical NER tasks and possibly reducing the reliance on extensive annotated datasets. OBJECTIVES: This study quantifies the capabilities of GPT-3.5 and GPT-4 for clinical named entity recognition (NER) tasks and proposes task-specific prompts to improve their performance. MATERIALS AND METHODS: We evaluated these models on 2 clinical NER tasks: (1) to extract medical problems, treatments, and tests from clinical notes in the MTSamples corpus, following the 2010 i2b2 concept extraction shared task, and (2) to identify nervous system disorder-related adverse events from safety reports in the vaccine adverse event reporting system (VAERS). To improve the GPT models' performance, we developed a clinical task-specific prompt framework that includes (1) baseline prompts with task description and format specification, (2) annotation guideline-based prompts, (3) error analysis-based instructions, and (4) annotated samples for few-shot learning. We assessed each prompt's effectiveness and compared the models to BioClinicalBERT. RESULTS: Using baseline prompts, GPT-3.5 and GPT-4 achieved relaxed F1 scores of 0.634, 0.804 for MTSamples and 0.301, 0.593 for VAERS. Additional prompt components consistently improved model performance. When all 4 components were used, GPT-3.5 and GPT-4 achieved relaxed F1 socres of 0.794, 0.861 for MTSamples and 0.676, 0.736 for VAERS, demonstrating the effectiveness of our prompt framework. Although these results trail BioClinicalBERT (F1 of 0.901 for the MTSamples dataset and 0.802 for the VAERS), it is very promising considering few training samples are needed. DISCUSSION: The study's findings suggest a promising direction in leveraging LLMs for clinical NER tasks. However, while the performance of GPT models improved with task-specific prompts, there's a need for further development and refinement. LLMs like GPT-4 show potential in achieving close performance to state-of-the-art models like BioClinicalBERT, but they still require careful prompt engineering and understanding of task-specific knowledge. The study also underscores the importance of evaluation schemas that accurately reflect the capabilities and performance of LLMs in clinical settings. CONCLUSION: While direct application of GPT models to clinical NER tasks falls short of optimal performance, our task-specific prompt framework, incorporating medical knowledge and training samples, significantly enhances GPT models' feasibility for potential clinical applications.

METTL3-mediated pre-miR-665/DLX3 m6A methylation facilitates the committed differentiation of stem cells from apical papilla.

Methyltransferase-like 3 (METTL3) is a crucial element of N6-methyladenosine (m6A) modifications and has been extensively studied for its involvement in diverse biological and pathological processes. In this study, we explored how METTL3 affects the differentiation of stem cells from the apical papilla (SCAPs) into odonto/osteoblastic lineages through gain- and loss-of-function experiments. The m6A modification levels were assessed using m6A dot blot and activity quantification experiments. In addition, we employed Me-RIP microarray experiments to identify specific targets modified by METTL3. Furthermore, we elucidated the molecular mechanism underlying METTL3 function through dual-luciferase reporter gene experiments and rescue experiments. Our findings indicated that METTL3+/- mice exhibited significant root dysplasia and increased bone loss. The m6A level and odonto/osteoblastic differentiation capacity were affected by the overexpression or inhibition of METTL3. This effect was attributed to the acceleration of pre-miR-665 degradation by METTL3-mediated m6A methylation in cooperation with the "reader" protein YTHDF2. Additionally, the targeting of distal-less homeobox 3 (DLX3) by miR-665 and the potential direct regulation of DLX3 expression by METTL3, mediated by the "reader" protein YTHDF1, were demonstrated. Overall, the METTL3/pre-miR-665/DLX3 pathway might provide a new target for SCAP-based tooth root/maxillofacial bone tissue regeneration.

RefAI: a GPT-powered retrieval-augmented generative tool for biomedical literature recommendation and summarization.

OBJECTIVES: Precise literature recommendation and summarization are crucial for biomedical professionals. While the latest iteration of generative pretrained transformer (GPT) incorporates 2 distinct modes-real-time search and pretrained model utilization-it encounters challenges in dealing with these tasks. Specifically, the real-time search can pinpoint some relevant articles but occasionally provides fabricated papers, whereas the pretrained model excels in generating well-structured summaries but struggles to cite specific sources. In response, this study introduces RefAI, an innovative retrieval-augmented generative tool designed to synergize the strengths of large language models (LLMs) while overcoming their limitations. MATERIALS AND METHODS: RefAI utilized PubMed for systematic literature retrieval, employed a novel multivariable algorithm for article recommendation, and leveraged GPT-4 turbo for summarization. Ten queries under 2 prevalent topics ("cancer immunotherapy and target therapy" and "LLMs in medicine") were chosen as use cases and 3 established counterparts (ChatGPT-4, ScholarAI, and Gemini) as our baselines. The evaluation was conducted by 10 domain experts through standard statistical analyses for performance comparison. RESULTS: The overall performance of RefAI surpassed that of the baselines across 5 evaluated dimensions-relevance and quality for literature recommendation, accuracy, comprehensiveness, and reference integration for summarization, with the majority exhibiting statistically significant improvements (P-values <.05). DISCUSSION: RefAI demonstrated substantial improvements in literature recommendation and summarization over existing tools, addressing issues like fabricated papers, metadata inaccuracies, restricted recommendations, and poor reference integration. CONCLUSION: By augmenting LLM with external resources and a novel ranking algorithm, RefAI is uniquely capable of recommending high-quality literature and generating well-structured summaries, holding the potential to meet the critical needs of biomedical professionals in navigating and synthesizing vast amounts of scientific literature.