Protein-protein interaction and site prediction using transfer learning.

The advanced language models have enabled us to recognize protein-protein interactions (PPIs) and interaction sites using protein sequences or structures. Here, we trained the MindSpore ProteinBERT (MP-BERT) model, a Bidirectional Encoder Representation from Transformers, using protein pairs as inputs, making it suitable for identifying PPIs and their respective interaction sites. The pretrained model (MP-BERT) was fine-tuned as MPB-PPI (MP-BERT on PPI) and demonstrated its superiority over the state-of-the-art models on diverse benchmark datasets for predicting PPIs. Moreover, the model's capability to recognize PPIs among various organisms was evaluated on multiple organisms. An amalgamated organism model was designed, exhibiting a high level of generalization across the majority of organisms and attaining an accuracy of 92.65%. The model was also customized to predict interaction site propensity by fine-tuning it with PPI site data as MPB-PPISP. Our method facilitates the prediction of both PPIs and their interaction sites, thereby illustrating the potency of transfer learning in dealing with the protein pair task.

Prolonged myelin deficits contribute to neuron loss and functional impairments after ischaemic stroke.

Ischaemic stroke causes neuron loss and long-term functional deficits. Unfortunately, effective approaches to preserving neurons and promoting functional recovery remain unavailable. Oligodendrocytes, the myelinating cells in the CNS, are susceptible to oxygen and nutrition deprivation and undergo degeneration after ischaemic stroke. Technically, new oligodendrocytes and myelin can be generated by the differentiation of oligodendrocyte precursor cells (OPCs). However, myelin dynamics and their functional significance after ischaemic stroke remain poorly understood. Here, we report numerous denuded axons accompanied by decreased neuron density in sections from ischaemic stroke lesions in human brain, suggesting that neuron loss correlates with myelin deficits in these lesions. To investigate the longitudinal changes in myelin dynamics after stroke, we labelled and traced pre-existing and newly-formed myelin, respectively, using cell-specific genetic approaches. Our results indicated massive oligodendrocyte death and myelin loss 2 weeks after stroke in the transient middle cerebral artery occlusion (tMCAO) mouse model. In contrast, myelin regeneration remained insufficient 4 and 8 weeks post-stroke. Notably, neuronal loss and functional impairments worsened in aged brains, and new myelin generation was diminished. To analyse the causal relationship between remyelination and neuron survival, we manipulated myelinogenesis by conditional deletion of Olig2 (a positive regulator) or muscarinic receptor 1 (M1R, a negative regulator) in OPCs. Deleting Olig2 inhibited remyelination, reducing neuron survival and functional recovery after tMCAO. Conversely, enhancing remyelination by M1R conditional knockout or treatment with the pro-myelination drug clemastine after tMCAO preserved white matter integrity and neuronal survival, accelerating functional recovery. Together, our findings demonstrate that enhancing myelinogenesis is a promising strategy to preserve neurons and promote functional recovery after ischaemic stroke.

Identification of the Acid-Sensitive Site Critical for Chloral Hydrate (CH) Activation of the Proton-Activated Chloride Channel.

The transmembrane protein TMEM206 was recently identified as the molecular basis of the extracellular proton-activated Cl- channel (PAC), which plays an essential role in neuronal death in ischemia-reperfusion. The PAC channel is activated by extracellular acid, but the proton-sensitive mechanism remains unclear, although different acid-sensitive pockets have been suggested based on the cryo-EM structure of the human PAC (hPAC) channel. In the present study, we firstly identified two acidic amino acid residues that removed the pH-dependent activation of the hPAC channel by neutralization all the conservative negative charged residues located in the extracellular domain of the hPAC channel and some positively charged residues at the hotspot combined with two-electrode voltage-clamp (TEVC) recording in the Xenopus oocytes system. Double-mutant cycle analysis and double cysteine mutant of these two residues proved that these two residues cooperatively form a proton-sensitive site. In addition, we found that chloral hydrate activates the hPAC channel depending on the normal pH sensitivity of the hPAC channel. Furthermore, the PAC channel knock-out (KO) male mice (C57BL/6J) resist chloral hydrate-induced sedation and hypnosis. Our study provides a molecular basis for understanding the proton-dependent activation mechanism of the hPAC channel and a novel drug target of chloral hydrate.SIGNIFICANCE STATEMENT Proton-activated Cl- channel (PAC) channels are widely distributed in the nervous system and play a vital pathophysiological role in ischemia and endosomal acidification. The main discovery of this paper is that we identified the proton activation mechanism of the human proton-activated chloride channel (hPAC). Intriguingly, we also found that anesthetic chloral hydrate can activate the hPAC channel in a pH-dependent manner. We found that the chloral hydrate activates the hPAC channel and needs the integrity of the pH-sensitive site. In addition, the PAC channel knock-out (KO) mice are resistant to chloral hydrate-induced anesthesia. The study on PAC channels' pH activation mechanism enables us to better understand PAC's biophysical mechanism and provides a novel target of chloral hydrate.

Mapping the neural mechanism that distinguishes between holistic thinking and analytic thinking.

Holistic and analytic thinking are two distinct modes of thinking used to interpret the world with relative preferences varying across cultures. While most research on these thinking styles has focused on behavioral and cognitive aspects, a few studies have utilized functional magnetic resonance imaging (fMRI) to explore the correlations between brain metrics and self-reported scale scores. Other fMRI studies used single holistic and analytic thinking tasks. As a single task may involve processing in spurious low-level regions, we used two different holistic and analytic thinking tasks, namely the frame-line task and the triad task, to seek convergent brain regions to distinguish holistic and analytic thinking using multivariate pattern analysis (MVPA). Results showed that brain regions fundamental to distinguish holistic and analytic thinking include the bilateral frontal lobes, bilateral parietal lobes, bilateral precentral and postcentral gyrus, bilateral supplementary motor areas, bilateral fusiform, bilateral insula, bilateral angular gyrus, left cuneus, and precuneus, left olfactory cortex, cingulate gyrus, right caudate and putamen. Our study maps brain regions that distinguish between holistic and analytic thinking and provides a new approach to explore the neural representation of cultural constructs. We provide initial evidence connecting culture-related brain regions with language function to explain the origins of cultural differences in cognitive styles.

The SFTSV Nonstructural Proteins Induce Autophagy to Promote Viral Replication via Interaction with Vimentin.

Severe fever with thrombocytopenia syndrome virus (SFTSV) is a newly identified phlebovirus associated with severe hemorrhagic fever in humans. Studies have shown that SFTSV nucleoprotein (N) induces BECN1-dependent autophagy to promote viral assembly and release. However, the function of other SFTSV proteins in regulating autophagy has not been reported. In this study, we identify SFTSV NSs, a nonstructural protein that forms viroplasm-like structures in the cytoplasm of infected cells as the virus component mediating SFTSV-induced autophagy. We found that SFTSV NSs-induced autophagy was inclusion body independent, and most phenuivirus NSs had autophagy-inducing effects. Unlike N protein-induced autophagy, SFTSV NSs was key in regulating autophagy by interacting with the host's vimentin in an inclusion body-independent manner. NSs interacted with vimentin and induced vimentin degradation through the K48-linked ubiquitin-proteasome pathway. This negatively regulating Beclin1-vimentin complex formed and promoted autophagy. Furthermore, we identified the NSs-binding domain of vimentin and found that overexpression of wild-type vimentin antagonized the induced effect of NSs on autophagy and inhibited viral replication, suggesting that vimentin is a potential antiviral target. The present study shows a novel mechanism through which SFTSV nonstructural protein activates autophagy, which provides new insights into the role of NSs in SFTSV infection and pathogenesis. IMPORTANCE Severe fever with thrombocytopenia syndrome virus (SFTSV) is a newly emerging tick-borne pathogen that causes multifunctional organ failure and even death in humans. As a housekeeping mechanism for cells to maintain steady state, autophagy plays a dual role in viral infection and the host's immune response. However, the relationship between SFTSV infection and autophagy has not been described in detail yet. Here, we demonstrated that SFTSV infection induced complete autophagic flux and facilitated viral proliferation. We also identified a key mechanism underlying NSs-induced autophagy, in which NSs interacted with vimentin to inhibit the formation of the Beclin1-vimentin complex and induced vimentin degradation through K48-linked ubiquitination modification. These findings may help us understand the new functions and mechanisms of NSs and may aid in the identification of new antiviral targets.

Predicting who has delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage using machine learning approach: a multicenter, retrospective cohort study.

BACKGROUND: Early prediction of delayed cerebral ischemia (DCI) is critical to improving the prognosis of aneurysmal subarachnoid hemorrhage (aSAH). Machine learning (ML) algorithms can learn from intricate information unbiasedly and facilitate the early identification of clinical outcomes. This study aimed to construct and compare the ability of different ML models to predict DCI after aSAH. Then, we identified and analyzed the essential risk of DCI occurrence by preoperative clinical scores and postoperative laboratory test results. METHODS: This was a multicenter, retrospective cohort study. A total of 1039 post-operation patients with aSAH were finally included from three hospitals in China. The training group contained 919 patients, and the test group comprised 120 patients. We used five popular machine-learning algorithms to construct the models. The area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, precision, and f1 score were used to evaluate and compare the five models. Finally, we performed a Shapley Additive exPlanations analysis for the model with the best performance and significance analysis for each feature. RESULTS: A total of 239 patients with aSAH (23.003%) developed DCI after the operation. Our results showed that in the test cohort, Random Forest (RF) had an AUC of 0.79, which was better than other models. The five most important features for predicting DCI in the RF model were the admitted modified Rankin Scale, D-Dimer, intracranial parenchymal hematoma, neutrophil/lymphocyte ratio, and Fisher score. Interestingly, clamping or embolization for the aneurysm treatment was the fourth button-down risk factor in the ML model. CONCLUSIONS: In this multicenter study, we compared five ML methods, among which RF performed the best in DCI prediction. In addition, the essential risks were identified to help clinicians monitor the patients at high risk for DCI more precisely and facilitate timely intervention.

Anticipating the transmissibility of the 2022 mpox outbreak.

An ongoing outbreak of monkeypox virus (MPXV) was first reported in the United Kingdom on 6 May 2022. As of 17 November, there had been a total of 80 221 confirmed MPXV cases in over 110 countries. Based on data reported between 6 May and 30 June 2022 in the United Kingdom, Spain, and Germany, we applied a deep learning approach using convolutional neural networks to evaluate the parameters of the 2022 MPXV outbreak. The basic reproduction number (R0 ) of MPXV was estimated to be 2.32 in the United Kingdom, which indicates the active diffusion of MPXV since the beginning of the outbreak. The data from Spain and Germany produced higher median R0 values of 2.42 and 2.88, respectively. Importantly, the estimated R0 of MPXV in the three countries tends to the previously calculated R0 of smallpox (3.50 to 6.00). Furthermore, the incubation (1/ε) and infectious (1/γ) period was predicted between 9 and 10 days and 4-5 days, respectively. The R0 value derived from MPXV is consistent with the significantly increasing number of cases, indicating the risk of a rapid spread of MPXV worldwide, which would provide important insights for the prevention and control of MPXV epidemic.

Robust deep learning-based PET prognostic imaging biomarker for DLBCL patients: a multicenter study.

OBJECTIVE: To develop and independently externally validate robust prognostic imaging biomarkers distilled from PET images using deep learning techniques for precise survival prediction in patients with diffuse large B cell lymphoma (DLBCL). METHODS: A total of 684 DLBCL patients from three independent medical centers were included in this retrospective study. Deep learning scores (DLS) were generated from PET images using deep convolutional neural network architecture known as VGG19 and DenseNet121. These DLSs were utilized to predict progression-free survival (PFS) and overall survival (OS). Furthermore, multiparametric models were designed based on results from the Cox proportional hazards model and assessed through calibration curves, concordance index (C-index), and decision curve analysis (DCA) in the training and validation cohorts. RESULTS: The DLSPFS and DLSOS exhibited significant associations with PFS and OS, respectively (P<0.05) in the training and validation cohorts. The multiparametric models that incorporated DLSs demonstrated superior efficacy in predicting PFS (C-index: 0.866) and OS (C-index: 0.835) compared to competing models in training cohorts. In external validation cohorts, the C-indices for PFS and OS were 0.760 and. 0.770 and 0.748 and 0.766, respectively, indicating the reliable validity of the multiparametric models. The calibration curves displayed good consistency, and the decision curve analysis (DCA) confirmed that the multiparametric models offered more net clinical benefits. CONCLUSIONS: The DLSs were identified as robust prognostic imaging biomarkers for survival in DLBCL patients. Moreover, the multiparametric models developed in this study exhibited promising potential in accurately stratifying patients based on their survival risk.

The FACT-targeted drug CBL0137 enhances the effects of rituximab to inhibit B-cell non-Hodgkin's lymphoma tumor growth by promoting apoptosis and autophagy.

BACKGROUND: Aggressive B-cell non-Hodgkin's lymphoma (B-NHL) patients often develop drug resistance and tumor recurrence after conventional immunochemotherapy, for which new treatments are needed. METHODS: We investigated the antitumor effects of CBL0137. In vitro, cell proliferation was assessed by CCK-8 and colony formation assay. Flow cytometry was performed to analyze cell cycle progression, apoptosis, mitochondrial depolarization, and reactive oxygen species (ROS) production. Autophagy was detected by transmission electron microscopy and mGFP-RFP-LC3 assay, while western blotting was employed to detect proteins involved in apoptosis and autophagy. RNA-sequencing was conducted to analyze the transcription perturbation after CBL0137 treatment in B-NHL cell lines. Finally, the efficacy and safety of CBL0137, rituximab, and their combination were tested in vivo. RESULTS: CBL0137, a small molecule anticancer agent that has significant antitumor effects in B-NHL. CBL0137 sequesters the FACT (facilitates chromatin transcription) complex from chromatin to produce cytotoxic effects in B-NHL cells. In addition, we discovered novel anticancer mechanisms of CBL0137. CBL0137 inhibited human B-NHL cell proliferation by inducing cell cycle arrest in S phase via the c-MYC/p53/p21 pathway. Furthermore, CBL0137 triggers ROS generation and induces apoptosis and autophagy in B-NHL cells through the ROS-mediated PI3K/Akt/mTOR and MAPK signaling pathways. Notably, a combination of CBL0137 and rituximab significantly suppressed B-NHL tumor growth in subcutaneous models, consistent with results at the cellular level in vitro. CONCLUSIONS: CBL0137 has potential as a novel approach for aggressive B-NHL, and its combination with rituximab can provide new therapeutic options for patients with aggressive B-NHL. Video Abstract.

Immune-induced pneumonia in patients with advanced solid tumors treated with immunotherapy: a real-world assessment.

Aim: To investigate the computed tomography (CT) and clinical characteristics of immunotherapy-induced pneumonitis (IIP) in patients with advanced solid tumors. Patients & methods: CT and clinical data of 254 patients with advanced solid tumors treated with immune checkpoint inhibitors in our hospital were collected retrospectively. Results: The incidences of IIP in patients with non-small-cell lung cancer, lymphoma and gastrointestinal tumors were 19% (19/100), 9.8% (6/61) and 6.2% (4/65), respectively. The median onset time for all 31 IIP patients was 44 days (interquartile range: 24-65). Most IIP patients (21/31) had grade 1-2 disease. Multifocal ground-glass opacities (seen in 21/31 patients) were the main CT findings of IIP. Conclusion: Patients should be alerted to the risk of IIP, an adverse reaction that has a relatively low incidence but which is sometimes life-threatening.

The study aimed to investigate the clinical and computed tomography (CT) features of immunotherapy-induced pneumonitis (IIP) in patients with advanced solid tumors. To describe these characteristics, clinical and CT information of 254 patients with advanced solid tumors who were treated with drugs called immune checkpoint inhibitors were collected. The incidences of IIP in patients with non-small-cell lung cancer, lymphoma and gastrointestinal tumors were 19% (19/100), 9.8% (6/61) and 6.2% (4/65), respectively. The median time taken to develop IIP for all 31 IIP patients was 44 days. Most IIP patients had mild or moderate (grade 1­2) disease. The main CT findings of IIP were abnormalities called multifocal ground-glass opacities (21/31). Most IIP patients can recover well after glucocorticoid discontinuation. This real-world study was done to raise physicians' awareness of the possible development of IIP, an adverse reaction with a relatively low incidence but which is sometimes life-threatening, to highlight the variety of CT manifestations, and to provide advice on regulating the timing and method of glucocorticoid therapy.

Employing Pd nanoparticles decorated on halloysite nanotube/carbon composite for electrochemical aptasensing of HER2 in breast cancer patients.

An effective biosensing platform is described based on halloysite nanotube/carbon composite decorated with Pd nanoparticles (HNT/C@Pd NPs). A novel electrochemical aptasensor was designed using the proposed nano-platform to determine human epidermal growth factor receptor 2 (HER2), a breast cancer biomarker. Inherently, aptasensing interfaces provide high sensitivity and selectivity for tumor markers owing to the high specific surface area of HNT/C and good conductivity stems from deposition of Pd NPs into HNT/C composite. With a correlation coefficient of 0.996, the electrochemical aptasensor demonstrated a wide linear range from 0.03 ng/mL to 9 ng/mL. The limit of detection (LOD) of the established assay was 8 pg/mL based on S/N = 3 method. Further, the designed biosensor demonstrated acceptable selectivity, good reproducibility, and high stability. The applicability of the impedimetric sensor in human serum samples was also examined and compared to enzyme-linked immunosorbent assay (ELISA) assay (p-value >0.05). Based on the results, it was found that the proposed methodology can be used in quantification of breast cancer markers for early diagnosis and treatment.

Biochemical responses of freshwater microalgae Chlorella sorokiniana to combined exposure of Zn(â ¡) and estrone with simultaneous pollutants removal.

With the development of livestock industry, contaminants such as divalent zinc ions (Zn (Ⅱ)) and estrone are often simultaneously detected in livestock wastewater. Nevertheless, the combined toxicity of these two pollutants on microalgae is still unclear. Moreover, microalgae have the potential for biosorption and bioaccumulation of heavy metals and organic compounds. Thus, this study investigated the joint effects of Zn (Ⅱ) and estrone on microalgae Chlorella sorokiniana, in terms of growth, photosynthetic activity and biomolecules, as well as pollutants removal by algae. Interestingly, a low Zn (Ⅱ) concentration promoted C. sorokiniana growth and photosynthetic activity, while the high concentration experienced inhibition. As the increase of estrone concentration, chlorophyll a content increased continuously to resist the environmental stress. Concurrently, the secretion of extracellular polysaccharides and proteins by algae increased with exposure to Zn (Ⅱ) and estrone, reducing toxicity of pollutants to microalgae. Reactive oxygen species and superoxide dismutase activity increased as the increase of pollutant concentration after 96 h cultivation, but high pollutant concentrations resulted in damage of cells, as proved by increased MDA content. Additionally, C. sorokiniana displayed remarkable removal efficiency for Zn (Ⅱ) and estrone, reaching up to 86.14% and 84.96% respectively. The study provides insights into the biochemical responses of microalgae to pollutants and highlights the potential of microalgae in pollutants removal.

Is nurses' clinical competence associated with their moral identity and injury?

BACKGROUND: The enhancement of nursing care quality is closely related to the clinical competence of nurses, making it a crucial component within health systems. OBJECTIVE: The present study investigated the relationship between nurses' clinical competence, moral identity, and moral injury during the COVID-19 outbreak. RESEARCH DESIGN: This cross-sectional study was carried out among frontline nurses, using the Moral Identity Questionnaire (MIQ), the Moral Injury Symptom Scale-Healthcare Professionals version (MISS-HP), and the Competency Inventory for Registered Nurse (CIRN) as data collection tools. PARTICIPANTS: and research context: The research population for this study consisted of all frontline nurses (n = 251) employed in a hospital in southern Iran. Sampling was conducted between May 1, 2021 and September 30, 2021, during the COVID-19 outbreak. ETHICAL CONSIDERATIONS: The present study received approval from the research ethics committee of Rafsanjan University of Medical Sciences, with project No. 99267 and code of ethics ID No. IR. RUMS.REC.1399.262, dated 15.02.2021. RESULTS: According to the study findings, 42.2% of the nurses demonstrated high clinical competence, while 51.4% exhibited moderate clinical competence. The results indicated a positive correlation between moral identity and clinical competence but a negative correlation between moral injury and clinical competence. Furthermore, the variables of moral identity and moral injury were found to predict 10% of the variance in clinical competence. CONCLUSION: According to the results, moral identity and moral injury had an impact on the clinical competence of nurses. Therefore, implementing a program aimed at enhancing moral identity and providing training strategies to address moral injury during crises like the COVID-19 pandemic can lead to improvements in nurses' clinical competence and the overall quality of care they provide.

Internal and external validation of machine learning-assisted prediction models for mechanical ventilation-associated severe acute kidney injury.

BACKGROUND: Currently, very few preventive or therapeutic strategies are used for mechanical ventilation (MV)-associated severe acute kidney injury (AKI). OBJECTIVES: We developed clinical prediction models to detect the onset of severe AKI in the first week of intensive care unit (ICU) stay during the initiation of MV. METHODS: A large ICU database Medical Information Mart for Intensive Care IV (MIMIC-IV) was analysed retrospectively. Data were collected from the clinical information recorded at the time of ICU admission and during the initial 12 h of MV. Using univariate and multivariate analyses, the predictors were selected successively. For model development, two machine learning algorithms were compared. The primary goal was to predict the development of AKI stage 2 or 3 (AKI-23) and AKI stage 3 (AKI-3) in the first week of patients' ICU stay after initial 12 h of MV. The developed models were externally validated using another multicentre ICU database (eICU Collaborative Research Database, eICU) and evaluated in various patient subpopulations. RESULTS: Models were developed using data from the development cohort (MIMIC-IV: 2008-2016; n = 3986); the random forest algorithm outperformed the logistic regression algorithm. In the internal (MIMIC-IV: 2017-2019; n = 1210) and external (eICU; n = 1494) validation cohorts, the incidences of AKI-23 were 154 (12.7%) and 119 (8.0%), respectively, with areas under the receiver operator characteristic curve of 0.78 (95% confidence interval [CI]: 0.74-0.82) and 0.80 (95% CI: 0.76-0.84); the incidences of AKI-3 were 81 (6.7%) and 67 (4.5%), with areas under the receiver operator characteristic curve of 0.81 (95% CI: 0.76-0.87) and 0.80 (95% CI: 0.73-0.86), respectively. CONCLUSIONS: Models driven by machine learning and based on routine clinical data may facilitate the early prediction of MV-associated severe AKI. The validated models can be found at: https://apoet.shinyapps.io/mv_aki_2021_v2/.

[Effect of pre-pregnancy body mass index on first-trimester pregnant women glucose and lipid metabolism].

OBJECTIVE: To examine the association of pre-pregnancy body mass index(BMI) and first-trimester glucose and lipid metabolism. METHODS: A total of 298 pregnant women with high risk factors for gestational diabetes in early pregnancy, with an average age of 32.24 years, 66.11% were primiparous and the average time for blood collection was 12.5 weeks, were collected from August 2021 to April 2022 at the Department of Nutrition, Haidian District Maternal and Child Health Hospital, Beijing. Pregnant women were divided into low weight group(n=15), normal weight group(n=181), overweight/obese group(n=102) according to their pre-pregnancy BMI. Fasting glucose, fasting insulin, lipid quadruple, C-reactive protein, leptin and adiponectin levels were compared among groups. RESULTS: (1)Triglyceride concentrations were significantly higher in women in the overweight/obese group(1.51 mmol/L vs.1.15 mmol/L) than in women in low weight group(P<0.01). HDL cholesterol(HDL-C) levels were lower than in women with low weight(1.64 mmol/L vs.1.95 mmol/L)(P<0.01). (2)Pre-pregnancy low weight women had reduced pancreatic ß-cell function in early pregnancy with the HOMA-ß index of 60.41%, and women in the overweight/obese group had heavier fasting insulin levels(7.86 vs.3.42 µU/mL) and insulin resistance(1.75 vs 0.74) in early pregnancy than low weight women(P<0.01). (3)Pre-pregnancy BMI was positively correlated with triglycerides, fasting insulin, C-reactive protein levels, and degree of insulin resistance in early pregnancy(r=0.30, 0.28, 0.45 and 0.45, P<0.01)and negatively correlated with HDL-C levels in early pregnancy(r=-0.29, P<0.01). CONCLUSION: Among pregnant women with risk factors for gestational diabetes, pre-pregnancy overweight and obesity are associated with glucose and lipid metabolism levels in early pregnancy.

Optimal PET-based radiomic signature construction based on the cross-combination method for predicting the survival of patients with diffuse large B-cell lymphoma.

PURPOSE: To develop and externally validate models incorporating a PET radiomics signature (R-signature) obtained by the cross-combination method for predicting the survival of patients with diffuse large B-cell lymphoma (DLBCL). METHODS: A total of 383 patients with DLBCL from two medical centres between 2011 and 2019 were included. The cross-combination method was used on three types of PET radiomics features from the training cohort to generate 49 feature selection-classification candidates based on 7 different machine learning models. The R-signature was then built by selecting the optimal candidates based on their progression-free survival (PFS) and overall survival (OS). Cox regression analysis was used to develop the survival prediction models. The calibration, discrimination, and clinical utility of the models were assessed and externally validated. RESULTS: The R-signatures determined by 12 and 31 radiomics features were significantly associated with PFS and OS, respectively (P<0.05). The combined models that incorporated R-signatures, metabolic metrics, and clinical risk factors exhibited significant prognostic superiority over the clinical models, PET-based models, and the National Comprehensive Cancer Network International Prognostic Index in terms of both PFS (C-index: 0.801 vs. 0.732 vs. 0.785 vs. 0.720, respectively) and OS (C-index: 0.807 vs. 0.740 vs. 0.773 vs. 0.726, respectively). For external validation, the C-indices were 0.758 vs. 0.621 vs. 0.732 vs. 0.673 and 0.794 vs. 0.696 vs. 0.781 vs. 0.708 in the PFS and OS analyses, respectively. The calibration curves showed good consistency, and the decision curve analysis supported the clinical utility of the combined model. CONCLUSION: The R-signature could be used as a survival predictor for DLBCL, and its combination with clinical factors may allow for accurate risk stratification.

Radiomics signature from [18F]FDG PET images for prognosis predication of primary gastrointestinal diffuse large B cell lymphoma.

OBJECTIVES: To investigate the prognostic value of PET radiomics feature in the prognosis of patients with primary gastrointestinal diffuse large B cell lymphoma (PGI-DLBCL) treated with R-CHOP-like regimen. METHODS: A total of 140 PGI-DLBCL patients who underwent pre-therapy [18F] FDG PET/CT were enrolled in this retrospective analysis. PET radiomics features obtained from patients in the training cohort were subjected to three machine learning methods and Pearson's correlation test for feature selection. Support vector machine (SVM) was used to build a radiomics signature classifier associated with progression-free survival (PFS) and overall survival (OS). A multivariate Cox proportional hazards regression model was established to predict survival outcomes. RESULTS: A total of 1421 PET radiomics features were extracted and reduced to 5 features to build a radiomics signature which was significantly associated with PFS and OS (p < 0.05). The combined model incorporating radiomics signatures, metabolic metrics, and clinical risk factors showed high C-indices in both the training (PFS: 0.825, OS: 0.834) and validation sets (PFS: 0.831, OS: 0.877). Decision curve analysis (DCA) demonstrated that the combined models achieved the most net benefit across a wider reasonable range of threshold probabilities for predicting PFS and OS. CONCLUSION: The newly developed radiomics signatures obtained by the ensemble strategy were independent predictors of PFS and OS for PGI-DLBCL patients. Moreover, the combined model with clinical and metabolic factors was able to predict patient prognosis and may enable personalized treatment decision-making. KEY POINTS: • Radiomics signatures generated from the optimal radiomics feature set from the [18F]FDG PET images can predict the survival of PGI-DLBCL patients. • The optimal radiomics feature set is constructed by integrating the feature selection outputs of LASSO, RF, Xgboost, and PC methods. • Combined models incorporating radiomics signatures from18F-FDG PET images, metabolic parameters, and clinical factors outperformed clinical models, and NCCN-IPI.

Deep learning-based tumour segmentation and total metabolic tumour volume prediction in the prognosis of diffuse large B-cell lymphoma patients in 3D FDG-PET images.

OBJECTIVES: To demonstrate the effectiveness of automatic segmentation of diffuse large B-cell lymphoma (DLBCL) in 3D FDG-PET scans using a deep learning approach and validate its value in prognosis in an external validation cohort. METHODS: Two PET datasets were retrospectively analysed: 297 patients from a local centre for training and 117 patients from an external centre for validation. A 3D U-Net architecture was trained on patches randomly sampled within the PET images. Segmentation performance was evaluated by six metrics, including the Dice similarity coefficient (DSC), Jaccard similarity coefficient (JSC), sensitivity (Se), positive predictive value (PPV), Hausdorff distance 95 (HD 95), and average symmetric surface distance (ASSD). Finally, the prognostic value of predictive total metabolic tumour volume (pTMTV) was validated in real clinical applications. RESULTS: The mean DSC, JSC, Se, PPV, HD 95, and ASSD (with standard deviation) for the validation cohort were 0.78 ± 0.25, 0.69 ± 0.26, 0.81 ± 0.27, 0.82 ± 0.25, 24.58 ± 35.18, and 4.46 ± 8.92, respectively. The mean ground truth TMTV (gtTMTV) and pTMTV were 276.6 ± 393.5 cm3 and 301.9 ± 510.5 cm3 in the validation cohort, respectively. Perfect homogeneity in the Bland-Altman analysis and a strong positive correlation in the linear regression analysis (R2 linear = 0.874, p < 0.001) were demonstrated between gtTMTV and pTMTV. pTMTV (≥ 201.2 cm3) (PFS: HR = 3.097, p = 0.001; OS: HR = 6.601, p < 0.001) was shown to be an independent factor of PFS and OS. CONCLUSIONS: The FCN model with a U-Net architecture can accurately segment lymphoma lesions and allow fully automatic assessment of TMTV on PET scans for DLBCL patients. Furthermore, pTMTV is an independent prognostic factor of survival in DLBCL patients. KEY POINTS: •The segmentation model based on a U-Net architecture shows high performance in the segmentation of DLBCL patients on FDG-PET images. •The proposed method can provide quantitative information as a predictive TMTV for predicting the prognosis of DLBCL patients.

MiR-135b Alleviates Airway Inflammation in Asthmatic Children and Experimental Mice with Asthma via Regulating CXCL12.

OBJECTIVE: To clarify the possible influence of miR-135b on CXCL12 and airway inflammation in children and experimental mice with asthma. METHODS: The expressions of miR-135b and CXCL12 were detected using quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR) in the serum of asthmatic children. Besides, the experimental asthmatic mice were established by aerosol inhalation of ovalbumin (OVA) followed by the treatment with agomiR-135b and antagomir-135b. Pathological changes of lung tissues were observed via HE staining and PAS staining. Besides, the airway hyperresponsiveness of mice was elevated and bronchoalveolar lavage fluid (BALF) was isolated for cell categorization and counting. The inflammatory cytokines in BALF were determined by enzyme-linked immunosorbent assay (ELISA), and the infiltration of Th17 cells in lung tissues was measured using flow cytometry. RESULTS: MiR-135b was downregulated and CXCL12 was upregulated in asthmatic children and mice. Overexpression of miR-135b may down-regulate CXCL12 expression in the lung of OVA mice, resulting in significant decreases in inflammatory infiltration, hyperplasia of goblet cell, airway hyperresponsiveness, cell quantity, as well as the quantity of eosinophilic granulocytes, neutrophils and lymphocytes in BALF. Also, the levels of inflammatory cytokines (IL-4, IL-5, IL-13 and IL-17) and the ratio of Th17 cells and IL-17 levels in lung tissues were decreased. However, miR-135b downregulation reversed these changes in OVA mice. CONCLUSION: MiR-135b may inhibit immune responses of Th17 cells to alleviate airway inflammation and hyperresponsiveness in asthma possibly by targeting CXCL12, showing the potential value in asthma treatment.

Proteomic Analysis Reveals the Vital Role of Synaptic Plasticity in the Pathogenesis of Temporal Lobe Epilepsy.

Temporal lobe epilepsy (TLE) is a chronic neurological disorder that is often resistant to antiepileptic drugs. The pathogenesis of TLE is extremely complicated and remains elusive. Understanding the molecular mechanisms underlying TLE is crucial for its diagnosis and treatment. In the present study, a lithium-pilocarpine-induced TLE model was employed to reveal the pathological changes of hippocampus in rats. Hippocampal samples were taken for proteomic analysis at 2 weeks after the onset of spontaneous seizure (a chronic stage of epileptogenesis). Isobaric tag for relative and absolute quantization (iTRAQ) coupled with liquid chromatography-tandem mass spectrometry (LC-MS/MS) technique was applied for proteomic analysis of hippocampus. A total of 4173 proteins were identified from the hippocampi of epileptic rats and its control, of which 27 differentially expressed proteins (DEPs) were obtained with a fold change > 1.5 and P < 0.05. Bioinformatics analysis indicated 27 DEPs were mainly enriched in "regulation of synaptic plasticity and structure" and "calmodulin-dependent protein kinase activity," which implicate synaptic remodeling may play a vital role in the pathogenesis of TLE. Consequently, the synaptic plasticity-related proteins and synaptic structure were investigated to verify it. It has been demonstrated that CaMKII-α, CaMKII-ß, and GFAP were significant upregulated coincidently with proteomic analysis in the hippocampus of TLE rats. Moreover, the increased dendritic spines and hippocampal sclerosis further proved that synaptic plasticity involves in the development of TLE. The present study may help to understand the molecular mechanisms underlying epileptogenesis and provide a basis for further studies on synaptic plasticity in TLE.