A Mitochondria-Targeting Water-Soluble Fluorescent Probe for Selective Detection of Glyoxal in Living Cells.

Glyoxal (GL) is a physiological reactive α-oxoaldehyde metabolite, produced by lipid peroxidation and autoxidation of glucose. In this work, a specific mitochondria-targeting fluorescent probe Z-GL for glyoxal has been developed by an introducing isopropyl group on the recognition site to tune the selectivity toward glyoxal. The probe showed high selectivity and sensitivity for glyoxal in an aqueous system. Importantly, the probe was able to visualize exogenous and endogenous glyoxal in living cells. Furthermore, the probe was mitochondria-targetable, and could be used for monitoring the level of intracellular glyoxal.

Cross-Linked Polyimide Aerogels with Excellent Thermal and Mechanical Properties.

With the increasing development of productivity, new materials that allow for the efficient use of energy are slowly becoming a sought-after goal, as well as a challenge that is currently being faced. For this reason, we have made aerogels as the target of our research and prepared different series (CLPI (1-5)) of cross-linked polyimide aerogels by mixing and cross-linking the heat-insulating cross-linking agent 1,3,5-tris(4-aminobenzylamino)benzene (TAB) with polyamic acid solution. We created a three-dimensional spatial organization by using vacuum freeze-drying and programmed high-temperature drying, then controlled the concentration of the polyamidate solution to investigate the concentration and TAB's influence on aerogel-related properties. Among them, the shrinkage is reduced from 40% in CLPI-1 to 28% in CLPI-5, and it also shows excellent mechanical characteristics, the highest compression strength (CLPI-5) reaches 0.81 MPa and specific modulus reaches 41.95 KN m/Kg. In addition, adding TAB improves the aerogel thermal resistance, T5 in N2 from PI-2 519 °C to CLPI-2 556 °C. The three-dimensional network-type structure of the aerogel shows an excellent thermal insulation effect, where the thermal conductivity can be as low as 24.4 mWm-1 K-1. Compared with some protective materials, cross-linked polyimide aerogel presents better flame-retardant properties, greatly improving the scope of its application in the industrial protection.

Graph based recurrent network for context specific synthetic lethality prediction.

The concept of synthetic lethality (SL) has been successfully used for targeted therapies. To further explore SL for cancer therapy, identifying more SL interactions with therapeutic potential are essential. Recently, graph neural network-based deep learning methods have been proposed for SL prediction, which reduce the SL search space of wet-lab based methods. However, these methods ignore that most SL interactions depend strongly on genetic context, which limits the application of the predicted results. In this study, we proposed a graph recurrent network-based model for specific context-dependent SL prediction (SLGRN). In particular, we introduced a Graph Recurrent Network-based encoder to acquire a context-specific, low-dimensional feature representation for each node, facilitating the prediction of novel SL. SLGRN leveraged gate recurrent unit (GRU) and it incorporated a context-dependent-level state to effectively integrate information from all nodes. As a result, SLGRN outperforms the state-of-the-arts models for SL prediction. We subsequently validate novel SL interactions under different contexts based on combination therapy or patient survival analysis. Through in vitro experiments and retrospective clinical analysis, we emphasize the potential clinical significance of this context-specific SL prediction model.

Prophylactic endotracheal intubation before endoscopic surgery reduces the rebleeding rate in acute esophagogastric variceal bleeding patients.

Objectives: Esophagogastric variceal bleeding (EVB) is one of the main causes of cirrhosis-related deaths, and endoscopic therapy is the first-line treatment of choice. However, the efficacy of prophylactic endotracheal intubation (PEI) before endoscopy remains controversial. Methods: Data were collected from 119 patients who underwent endoscopic confirmation of an EVB. Inverse probability of treatment weighting was applied to reduce bias between the two groups. The primary outcomes included rebleeding rates within 24 h and 6 weeks post-endoscopic surgery and 6-week mortality. Results: After endoscopic surgery, the rebleeding rate within 24 h in the PEI group was significantly lower than non-PEI group (1.2 % VS 12.6 %, P-value = 0.025). Although PEI did not reduce 6-week mortality, it significantly reduced the risk of rebleeding within 24 h (odds ratio [OR]: 0.89, 95 % confidence interval [CI]: 0.82-0.97, P = 0.008) and within 6 weeks (hazard ratio [HR]: 0.36, 95%CI: 0.14-0.90, P = 0.029). In multivariate regression analyses, maximum varices diameter >1.5 cm (OR: 1.23, 95 % CI: 1.09-1.37, P < 0.001) was independent risk factor for rebleeding within 24 h. Creatinine (HR: 1.01, 95 % CI: 1.01-1.02, P < 0.001) and international normalized ratio (HR: 2.99, 95 % CI: 1.99-4.65, P < 0.001) were independent risk factors for rebleeding within 6 weeks. Conclusions: PEI before endoscopic surgery reduced the incidence of rebleeding within 24 h and 6 weeks after endoscopic surgery. However, PEI did not reduce the 6-week mortality rate after endoscopic surgery and might increase the length of hospital stay.

Loss of Carbamoyl Phosphate Synthetase 1 Potentiates Hepatocellular Carcinoma Metastasis by Reducing Aspartate Level.

Hepatocellular carcinoma (HCC) is one of the most lethal cancers worldwide. Numerous studies have shown that metabolic reprogramming is crucial for the development of HCC. Carbamoyl phosphate synthase 1 (CPS1), a rate-limiting enzyme in urea cycle, is an abundant protein in normal hepatocytes, however, lacking systemic research in HCC. It is found that CPS1 is low-expressed in HCC tissues and circulating tumor cells, negatively correlated with HCC stage and prognosis. Further study reveals that CPS1 is a double-edged sword. On the one hand, it inhibits the activity of phosphatidylcholine-specific phospholipase C to block the biosynthesis of diacylglycerol (DAG), leading to the downregulation of the DAG/protein kinase C pathway to inhibit invasion and metastasis of cancer cells. On the other hand, CPS1 promotes cell proliferation by increasing intracellular S-adenosylmethionin to enhance the m6A modification of solute carrier family 1 member 3 mRNA, a key transporter for aspartate intake. Finally, CPS1 overexpressing adeno-associated virus can dampen HCC progression. Collectively, this results uncovered that CPS1 is a switch between HCC proliferation and metastasis by increasing intracellular aspartate level.

A Multi-task Neural Network for Image Recognition in Magnetically Controlled Capsule Endoscopy.

BACKGROUND AND AIMS: Physicians are required to spend a significant amount of reading time of magnetically controlled capsule endoscopy. However, current deep learning models are limited to completing a single recognition task and cannot replicate the diagnostic process of a physician. This study aims to construct a multi-task model that can simultaneously recognize gastric anatomical sites and gastric lesions. METHODS: A multi-task recognition model named Mul-Recog-Model was established. The capsule endoscopy image data from 886 patients were selected to construct a training set and a test set for training and testing the model. Based on the same test set, the model in this study was compared with the current single-task recognition model with good performance. RESULTS: The sensitivity and specificity of the model for recognizing gastric anatomical sites were 99.8% (95% confidence intervals: 99.7-99.8) and 98.5% (95% confidence intervals: 98.3-98.7), and for gastric lesions were 98.8% (95% confidence intervals: 98.3-99.2) and 99.4% (95% confidence intervals: 99.1-99.7). Moreover, the positive predictive value, negative predictive value, and accuracy of the model were more than 95% in recognizing gastric anatomical sites and gastric lesions. Compared with the current single-task recognition model, our model showed comparable sensitivity, specificity, positive predictive value, negative predictive value, and accuracy (p < 0.01, except for the negative predictive value of ResNet, p > 0.05). The Areas Under Curve of our model were 0.985 and 0.989 in recognizing gastric anatomical sites and gastric lesions. Furthermore, the model had 49.1 M parameters and 38.1G Float calculations. The model took 15.5 ms to recognize an image, which was less than the superposition of multiple single models (p < 0.01). CONCLUSIONS: The Mul-Recog-Model exhibited high sensitivity, specificity, PPV, NPV, and accuracy. The model demonstrated excellent performance in terms of parameters quantity, Float computation, and computing time. The utilization of the model for recognizing gastric images can improve the efficiency of physicians' reports and meet complex diagnostic requirements.

Nitrate sources and transformation in surface water and groundwater in Huazhou District, Shaanxi, China: Integrated research using hydrochemistry, isotopes and MixSIAR model.

Global water resources affected by excessive nitrate (NO3-) have caused a series of human health and ecological problems. Therefore, identification of NO3- sources and transformations is of pivotal significance in the strategic governance of widespread NO3- contaminant. In this investigation, a combination of statistical analysis, chemical indicators, isotopes, and MixSIAR model approaches was adopted to reveal the hydrochemical factors affecting NO3- concentrations and quantify the contribution of each source to NO3- concentrations in surface water and groundwater. The findings revealed that high groundwater NO3- concentration is concentrated in the southwestern region, peaking at 271 mg/L. NO3- concentration in the Wei River and Yuxian River exhibited an increase from upstream to downstream, but in the Shidi River and Luowen River, its concentration was highest in the upstream. Groundwater NO3- has noticeable correlation with Na+, Ca2+, Mg2+, Cl-, HCO3-, TDS, EC, and ORP. In surface water, NO3- level is significantly correlated with NH4+ and ORP. Major sources of NO3- in surface and groundwater comprise manure & sewage and soil nitrogen. Source contribution for surface water was calculated by MixSIAR model to obtain soil nitrogen (57.7%), manure & sewage (23.8%), chemical fertilizer (12%), and atmospheric deposition (6.4%). In groundwater, soil nitrogen and manure & sewage accounted for 19% and 63.8% of nitrate sources, respectively. Both surface water and groundwater exhibited strong oxidation, with nitrification the primary process. It is expected that this study will provide insights into the dynamics of NO3- and contribute to the development of effective strategies for mitigating NO3- contaminant, leading to sustainable management of water resources.

Geometric Formulation of Generalized Root-TT[over ¯] Deformations.

We develop a generic geometric formalism that incorporates both TT[over ¯]-like and root-TT[over ¯]-like deformations in arbitrary dimensions. This framework applies to a wide family of stress-energy tensor perturbations and encompasses various well-known field theories. Building upon the recently proposed correspondence between Ricci-based gravity and TT[over ¯]-like deformations, we further extend this duality to include root-TT[over ¯]-like perturbations. This refinement extends the potential applications of our approach and contributes to a deeper exploration of the interplay between stress tensor perturbations and gravitational dynamics. Among the various original outcomes detailed in this Letter, we have also obtained a deformation of the flat Jackiw-Teitelboim gravity action.

A Point Cloud Graph Neural Network for Protein-Ligand Binding Site Prediction.

Predicting protein-ligand binding sites is an integral part of structural biology and drug design. A comprehensive understanding of these binding sites is essential for advancing drug innovation, elucidating mechanisms of biological function, and exploring the nature of disease. However, accurately identifying protein-ligand binding sites remains a challenging task. To address this, we propose PGpocket, a geometric deep learning-based framework to improve protein-ligand binding site prediction. Initially, the protein surface is converted into a point cloud, and then the geometric and chemical properties of each point are calculated. Subsequently, the point cloud graph is constructed based on the inter-point distances, and the point cloud graph neural network (GNN) is applied to extract and analyze the protein surface information to predict potential binding sites. PGpocket is trained on the scPDB dataset, and its performance is verified on two independent test sets, Coach420 and HOLO4K. The results show that PGpocket achieves a 58% success rate on the Coach420 dataset and a 56% success rate on the HOLO4K dataset. These results surpass competing algorithms, demonstrating PGpocket's advancement and practicality for protein-ligand binding site prediction.

An attentional mechanism model for segmenting multiple lesion regions in the diabetic retina.

Diabetic retinopathy (DR), a leading cause of blindness in diabetic patients, necessitates the precise segmentation of lesions for the effective grading of lesions. DR multi-lesion segmentation faces the main concerns as follows. On the one hand, retinal lesions vary in location, shape, and size. On the other hand, the currently available multi-lesion region segmentation models are insufficient in their extraction of minute features and are prone to overlooking microaneurysms. To solve the above problems, we propose a novel deep learning method: the Multi-Scale Spatial Attention Gate (MSAG) mechanism network. The model inputs images of varying scales in order to extract a range of semantic information. Our innovative Spatial Attention Gate merges low-level spatial details with high-level semantic content, assigning hierarchical attention weights for accurate segmentation. The incorporation of the modified spatial attention gate in the inference stage enhances precision by combining prediction scales hierarchically, thereby improving segmentation accuracy without increasing the associated training costs. We conduct the experiments on the public datasets IDRiD and DDR, and the experimental results show that the proposed method achieves better performance than other methods.

Dose escalation in radical radio(chemo)therapy for cervical and upper thoracic esophageal cancer with 3DCRT/IMRT (ChC&UES): a multicenter retrospective study.

BACKGROUND: Cervical and upper thoracic esophageal cancer (ESCA) presents treatment challenges due to limited clinical evidence. This multi-center study (ChC&UES) explores radical radio(chemo)therapy efficacy and safety, especially focusing on radiation dose. METHOD: We retrospectively analyzed clinical data from 1,422 cases across 8 medical centers. According to the radiation dose for primary gross tumor, patients were divided into standard dose radiotherapy (SD, 50-55 Gy) or high dose (HD, > 55 Gy) radiotherapy. HD was further subdivided into conventional- high-dose group (HD-conventional, 55-63 Gy) and ultra-high-dose group (HD-ultra, ≥ 63 Gy). Primary outcome was Overall Survival (OS). RESULTS: The median OS was 33.0 months (95% CI: 29.401-36.521) in the whole cohort. Compared with SD, HD shown significant improved survival in cervical ESCA in Kaplan-Meier (P = 0.029) and cox multivariate regression analysis (P = 0.024) while shown comparable survival in upper thoracic ESCA (P = 0.735). No significant difference existed between HD-conventional and HD-ultra in cervical (P = 0.976) and upper thoracic (P = 0.610) ESCA. Incidences of radiation esophagitis and pneumonia from HD were comparable to SD (P = 0.097, 0.240), while myosuppression risk was higher(P = 0.039). The Bonferroni method revealed that, for both cervical and upper thoracic ESCA, HD-ultra enhance the objective response rate (ORR) compared to SD (P < 0.05). CONCLUSION: HD radiotherapy benefits cervical but not upper thoracic ESCA, while increasing bone marrow suppression risk. Further dose escalating (≥ 63 Gy) doesn't improve survival but enhances ORR.

Deubiquitinase USP18 inhibits hepatic stellate cells activation and alleviates liver fibrosis via regulation of TAK1 activity.

BACKGROUND & AIMS: Activation of hepatic stellate cells (HSCs) is the key process underlying liver fibrosis. Unveiling its molecular mechanism may provide an effective target for inhibiting liver fibrosis. Protein ubiquitination is a dynamic and reversible process. Deubiquitinases (DUBs) catalyze the removal of ubiquitin chains from substrate proteins, thereby inhibiting the biological processes regulated by ubiquitination signals. However, there are few studies revealing the role of deubiquitination in the activation of HSCs. METHODS & RESULTS: Single-cell RNA sequencing (scRNA-seq) revealed significantly decreased USP18 expression in activated HSCs when compared to quiescent HSCs. In mouse primary HSCs, continuous activation of HSCs led to a gradual decrease in USP18 expression whilst restoration of USP18 expression significantly inhibited HSC activation. Injection of USP18 lentivirus into the portal vein of a CCl4-induced liver fibrosis mouse model confirmed that overexpression of USP18 can significantly reduce the degree of liver fibrosis. In terms of mechanism, we screened some targets of USP18 in mouse primary HSCs and found that USP18 could directly bind to TAK1. Furthermore, we demonstrated that USP18 can inhibit TAK1 activity by interfering with the K63 ubiquitination of TAK1. CONCLUSIONS: Our study demonstrated that USP18 inhibited HSC activation and alleviated liver fibrosis via modulation of TAK1 activity; this may prove to be an effective target for inhibiting liver fibrosis.

Genome-wide sRNA and mRNA transcriptomic profiling insights into carbapenem-resistant Acinetobacter baumannii.

Introduction: Acinetobacter baumannii (AB) is rising as a human pathogen of critical priority worldwide as it is the leading cause of opportunistic infections in healthcare settings and carbapenem-resistant AB is listed as a "super bacterium" or "priority pathogen for drug resistance" by the World Health Organization. Methods: Clinical isolates of A. baumannii were collected and tested for antimicrobial susceptibility. Among them, carbapenem-resistant and carbapenem-sensitive A. baumannii were subjected to prokaryotic transcriptome sequencing. The change of sRNA and mRNA expression was analyzed by bioinformatics and validated by quantitative reverse transcription-PCR. Results: A total of 687 clinical isolates were collected, of which 336 strains of A. baumannii were resistant to carbapenem. Five hundred and six differentially expressed genes and nineteen differentially expressed sRNA candidates were discovered through transcriptomic profile analysis between carbapenem-resistant isolates and carbapenem-sensitive isolates. Possible binding sites were predicted through software for sRNA21 and adeK, sRNA27 and pgaC, sRNA29 and adeB, sRNA36 and katG, indicating a possible targeting relationship. A negative correlation was shown between sRNA21 and adeK (r = -0.581, P = 0.007), sRNA27 and pgaC (r = -0.612, P = 0.004), sRNA29 and adeB (r = -0.516, P = 0.020). Discussion: This study preliminarily screened differentially expressed mRNA and sRNA in carbapenem-resistant A. baumannii, and explored possible targeting relationships, which will help further reveal the resistance mechanism and provide a theoretical basis for the development of drugs targeting sRNA for the prevention and treatment of carbapenem-resistant A. baumannii infection.

[Causes of missed diagnosis of clinically significant prostate cancer by targeted biopsy].

OBJECTIVE: To retrospectively analyze the causes of missed diagnosis of clinically significant PCa (csPCa) by targeted biopsy (TB). METHODS: This retrospective study included 652 males aged (71.32 ± 16.53) years with elevated PSA and abnormal MRI signals detected in our hospital from June 2018 to December 2020. We further examined the patients by transperineal prostatic TB and systematic biopsy (SB), analyzed the detection rates of PCa and csPCa by TB and SB, and investigated the causes of missed diagnosis of csPCa in TB using the fishbone diagram. RESULTS: The total detection rate of PCa and csPCa by TB combined with SB was 45.7% (298/652), and that of csPCa was 37.4% (244/652), with 38 cases of csPCa missed in TB, including 23 cases of negative TB and 15 cases of low ISUP grade. The causes of missed diagnosis of csPCa by TB included low MRI image quality, PSA density ≤0.15 ng/ml/cm3, target area <10 mm, and PI-RADS 2 score ≤3. The detection rate of csPCa by TB alone was 31.6%, which was increased by 5.8% (P = 0.027) when TB combined with SB. CONCLUSION: TB combined with SB yields a higher detection rate of csPCa than either used alone. Missed diagnosis of csPCa by TB is closely related to the characteristics of tumor and MR image of the target area.

Mito-LND and (E)-Akt inhibitor-IV: novel compounds inducing endoplasmic reticulum stress and ROS accumulation against hepatocellular carcinoma.

BACKGROUND: Hepatocellular carcinoma (HCC) is one of the leading causes of cancer-related mortality. Although multi-kinase inhibitors can prolong the overall survival of late-stage HCC patients, the emergence of drug resistance diminishes these benefits, ultimately resulting in treatment failure. Therefore, there is an urgent need for novel and effective drugs to impede the progression of liver cancer. METHODS: This study employed a concentration gradient increment method to establish acquired sorafenib or regorafenib-resistant SNU-449 cells. Cell viability was assessed using the cell counting kit-8 assay. A library of 793 bioactive small molecules related to metabolism screened compounds targeting both parental and drug-resistant cells. The screened compounds will be added to both the HCC parental cells and the drug-resistant cells, followed by a comprehensive assessment. Intracellular adenosine triphosphate (ATP) levels were quantified using kits. Flow cytometry was applied to assess cell apoptosis and reactive oxygen species (ROS). Real-time quantitative PCR studied relative gene expression, and western blot analysis assessed protein expression changes in HCC parental and drug-resistant cells. A xenograft model in vivo evaluated Mito-LND and (E)-Akt inhibitor-IV effects on liver tumors, with hematoxylin and eosin staining for tissue structure and immunohistochemistry staining for endoplasmic reticulum stress protein expression. RESULTS: From the compound library, we screened out two novel compounds, Mito-LND and (E)-Akt inhibitor-IV, which could potently kill both parental cells and drug-resistant cells. Mito-LND could significantly suppress proliferation and induce apoptosis in HCC parental and drug-resistant cells by upregulating glycolytic intermediates and downregulating those of the tricarboxylic acid (TCA) cycle, thereby decreasing ATP production and increasing ROS. (E)-Akt inhibitor-IV achieved comparable results by reducing glycolytic intermediates, increasing TCA cycle intermediates, and decreasing ATP synthesis and ROS levels. Both compounds trigger apoptosis in HCC cells through the interplay of the AMPK/MAPK pathway and the endoplasmic reticulum stress response. In vivo assays also showed that these two compounds could significantly inhibit the growth of HCC cells and induce endoplasmic reticulum stress. CONCLUSION: Through high throughput screening, we identified that Mito-LND and (E)-Akt inhibitor-IV are two novel compounds against both parental and drug-resistant HCC cells, which could offer new strategies for HCC patients.

Carbon capture in power sector of China towards carbon neutrality and its comparison to renewable power.

CO2 capture from coal power plants is an important and necessary solution to realizing carbon neutrality in China, but CCS demonstration deployment in power sector is far behind expectations. Hence, the reduction potential of energy consumption and cost for CCS and its competitiveness to renewable powers are very important to make roadmaps and policies toward carbon neutrality. Unlike the popular recognition that capturing CO2 from flue gases is technically and commercially mature, this paper notes that it has been proved to be technically feasible but far beyond technology maturity and high energy penalty leads to its immaturity and therefore causes high cost. Additionally, the potential energy penalty reduction of capture is investigated thermodynamically, and future CO2 avoidance cost is predicted and compared to renewable power (solar PV and onshore wind power). Results show that energy penalty for CO2 capture can be reduced by 48%-57%. When installation capacity reaches a similar scale to that of solar PV in China (250 GW), CO2 capture cost in coal power plants can be reduced from the current 28-40 US$/ton to 10-20 US$/ton, and efficiency upgrade contributes to 67%-75% in cost reduction for high coal price conditions. In China, CO2 capture in coal power plants can be cost competitive with solar PV and onshore wind power. But it is worth noting that the importance and share of CCS role in CO2 emission reduction is decreasing since renewable power is already well deployed and there is still a lack of large-scale CO2 capture demonstrations in China. Innovative capture technologies with low energy penalties need to be developed to promote CCS. Results in this work can provide informative references for making roadmaps and policies regarding CO2 emission reductions that contribute towards carbon neutrality.

Selectively lighting up glyoxal in living cells using an o-phenylenediamine fused hemicyanine.

Glyoxal (GL) is a reactive α-dicarbonyl compound generated from glycated proteins in the Maillard reaction. It has attracted particular attention over the past few years because of its possible clinical significance in chronic and age-related diseases. In this work, a reaction-based red emission fluorescent probe GL1 has been synthesized successfully by grafting an alkyl group onto an amino group to regulate its selectivity for GL. Under physiological conditions, the fluorescence intensity of GL1 at 640 nm obviously increased with the increase of GL concentration, and it exhibited high selectivity for GL over other reactive carbonyl compounds, as well as a lower detection limit (0.021 µM) and a larger Stokes shift (112 nm). At the same time, GL1 can selectively accumulate in mitochondria and can be used to detect exogenous and endogenous GL in living cells with low cytotoxicity.

Computational identification of novel potential genetic pathogenesis and otherwise biomarkers in acute liver allograft rejection.

Acute cellular rejection (ACR) is a prevalent postoperative complication following liver transplantation (LT), exhibiting an increasing incidence of morbidity and mortality. However, the molecular mechanisms of ACR following LT remain unclear. To explore the genetic pathogenesis and identify biomarkers of ACR following LT, three relevant Gene Expression Omnibus (GEO) datasets consisting of data on ACR or non-ACR patients after LT were comprehensively investigated by computational analysis. A total of 349 upregulated and 260 downregulated differentially expressed genes (DEGs) and eight hub genes (ISG15, HELZ2, HNRNPK, TIAL1, SKIV2L2, PABPC1, SIRT1, and PPARA) were identified. Notably, HNRNPK, TIAL1, and PABPC1 exhibited the highest predictive potential for ACR with AUCs of 0.706, 0.798, and 0.801, respectively. KEGG analysis of hub genes revealed that ACR following LT was predominately associated with ferroptosis, protein processing in the endoplasmic reticulum, complement and coagulation pathways, and RIG-I/NOD/Toll-like receptor signaling pathway. According to the immune cell infiltration analysis, γδT cells, NK cells, Tregs, and M1/M2-like macrophages had the highest levels of infiltration. Compared to SIRT1, ISG15 was positively correlated with γδT cells and M1-like macrophages but negatively correlated with NK cells, CD4+ memory T cells, and Tregs. In conclusion, this study identified eight hub genes and their potential pathways, as well as the immune cells involved in ACR following LT with the greatest levels of infiltration. These findings provide a new direction for future research on the underlying mechanism of ACR following LT.

Deep active learning with high structural discriminability for molecular mutagenicity prediction.

The assessment of mutagenicity is essential in drug discovery, as it may lead to cancer and germ cells damage. Although in silico methods have been proposed for mutagenicity prediction, their performance is hindered by the scarcity of labeled molecules. However, experimental mutagenicity testing can be time-consuming and costly. One solution to reduce the annotation cost is active learning, where the algorithm actively selects the most valuable molecules from a vast chemical space and presents them to the oracle (e.g., a human expert) for annotation, thereby rapidly improving the model's predictive performance with a smaller annotation cost. In this paper, we propose muTOX-AL, a deep active learning framework, which can actively explore the chemical space and identify the most valuable molecules, resulting in competitive performance with a small number of labeled samples. The experimental results show that, compared to the random sampling strategy, muTOX-AL can reduce the number of training molecules by about 57%. Additionally, muTOX-AL exhibits outstanding molecular structural discriminability, allowing it to pick molecules with high structural similarity but opposite properties.

Odor Fingerprinting of Chitosan and Source Identification of Commercial Chitosan: HS-GC-IMS, Multivariate Statistical Analysis, and Tracing Path Study.

Chitosan samples were prepared from the shells of marine animals (crab and shrimp) and the cell walls of fungi (agaricus bisporus and aspergillus niger). Fourier-transform infrared spectroscopy (FT-IR) was used to detect their molecular structures, while headspace-gas chromatography-ion mobility spectrometry (HS-GC-IMS) was employed to analyze their odor composition. A total of 220 volatile organic compounds (VOCs), including esters, ketones, aldehydes, etc., were identified as the odor fingerprinting components of chitosan for the first time. A principal component analysis (PCA) revealed that chitosan could be effectively identified and classified based on its characteristic VOCs. The sum of the first three principal components explained 87% of the total variance in original information. An orthogonal partial least squares discrimination analysis (OPLS-DA) model was established for tracing and source identification purposes, demonstrating excellent performance with fitting indices R2X = 0.866, R2Y = 0.996, Q2 = 0.989 for independent variable fitting and model prediction accuracy, respectively. By utilizing OPLS-DA modeling along with a heatmap-based tracing path study, it was found that 29 VOCs significantly contributed to marine chitosan at a significance level of VIP > 1.00 (p < 0.05), whereas another set of 20 VOCs specifically associated with fungi chitosan exhibited notable contributions to its odor profile. These findings present a novel method for identifying commercial chitosan sources, which can be applied to ensure biological safety in practical applications.