Metagenomic perspectives on antibiotic resistance genes in tap water: The environmental characteristic, potential mobility and health threat.

As an emerging environmental contaminant, antibiotic resistance genes (ARGs) in tap water have attracted great attention. Although studies have provided ARG profiles in tap water, research on their abundance levels, composition characteristics, and potential threat is still insufficient. Here, 9 household tap water samples were collected from the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) in China. Additionally, 75 sets of environmental sample data (9 types) were downloaded from the public database. Metagenomics was then performed to explore the differences in the abundance and composition of ARGs. 221 ARG subtypes consisting of 17 types were detected in tap water. Although the ARG abundance in tap water was not significantly different from that found in drinking water plants and reservoirs, their composition varied. In tap water samples, the three most abundant classes of resistance genes were multidrug, fosfomycin and MLS (macrolide-lincosamide-streptogramin) ARGs, and their corresponding subtypes ompR, fosX and macB were also the most abundant ARG subtypes. Regarding the potential mobility, vanS had the highest abundance on plasmids and viruses, but the absence of key genes rendered resistance to vancomycin ineffective. Generally, the majority of ARGs present in tap water were those that have not been assessed and are currently not listed as high-threat level ARG families based on the World Health Organization Guideline. Although the current potential threat to human health posed by ARGs in tap water is limited, with persistent transfer and accumulation, especially in pathogens, the potential danger to human health posed by ARGs should not be ignored.

An intelligent fluorescence sensing platform based on entropy-driven toehold-mediated strand displacement cycle reaction for point-of-care testing of miRNA.

BACKGROUND: MicroRNA (miRNA) has gradually become an emerging biomarker for early diagnosis and prognosis of various diseases due to its specific gene expression and high stability. With the development of molecular diagnosis and point-of-care testing (POCT) technology, developing simple, fast, sensitive, efficient, and low-cost miRNA sensors is of great significance for clinical applications and emergency rapid diagnosis. At present, entropy-driven toehold mediated chain displacement reaction, as a promising enzyme free isothermal amplification technique, is an important tool for ultra-sensitive biosensing applications. RESULTS: In this study, we used gold nanoparticles (AuNPs) as carriers and quenchers, modified them using self-assembled triple chain composite substrates AuNPs@A@B1/B2, and used dual reporter molecules for cascade cyclic amplification to amplify fluorescence signals, which proposed a fluorescent biosensor based on this reaction and build an intelligent fluorescence sensing platform for rapid detection of miRNA. We designed a highly specific self-programmable sensor using the acute ischemic stroke (AIS) biomarker miRNA-125a-5p as a sample, and achieved sensitive detection of miRNA in the range of 0.01 µM∼10 µ M under optimal conditions. It broke through the traditional detection limitations of weak signals and liberated the fluorescence detection environment. SIGNIFICANCE: In summary, this creative miRNA biosensor combined with POCT has demonstrated extraordinary detection potential, broad application prospects in the early diagnosis and prognosis monitoring of AIS, provides a novel miRNA universal detection strategy for the fields of biological and life sciences.

Petroleum hydrocarbons and colored dissolved organic matter shape marine oil-degrading microbiota in different patterns.

Both petroleum hydrocarbons (PHCs) from oil pollution and colored dissolved organic matter (CDOM) have great influences on the marine microbial community as carbon source factors. However, their combined effects and the specific influence patterns have been kept unclear. This study selected the northeastern South China Sea (NSCS), a typical oil contaminated area, and investigated the characteristics of oil-degrading microbiota in the seawaters by high-throughput sequencing and the relationships with PHCs and CDOM as well as other environmental factors. The results showed the oil pollution had induced the enrichment of oil-degrading bacteria and oil-degrading functional genes, resulting in the core function of oil-degrading microbiota for shaping the microbial community. The Mantel test indicated carbon source factors played the dominant role in shaping the oil-degrading microbiota, compared with geographical distance and other non­carbon source factors. The influence patterns and strength of PHCs and CDOM on oil-degrading microbiota were further comprehensively analyzed. PHCs played a driving role in the differentiation of oil-degrading microbiota, while CDOM played a stabilizing role for the community similarity. The constructed structural equation model confirmed their distinct influence patterns and also explored the mediating effects of bulk organic carbon. This work not only revealed the important impact of oil pollution on marine microbial communities, but also made people realize the self-regulation ability of the marine environment through the endogenous organic matter.

Recovery From Severe Mental Illnesses: The Influence of Personal and Psychosocial Factors in Community Settings.

Severe mental illnesses (SMIs) represent a significant public health challenge with substantial personal, economic and societal burdens. Despite treatment advances, recovery outcomes for SMI patients remain variable. This study aimed to investigate the influence of personal and psychosocial factors on recovery among SMI patients in community health settings. This cross-sectional study was conducted in 23 community health service centres in Nanjing, China. We recruited 924 participants diagnosed with SMIs, using a random sampling method. The relationships between individual, psychosocial factors and patients' recovery status were analysed by multiple linear regression. The mediation effect of medication adherence was further tested by the Sobel test with bootstrapping. The mean score of the Morningside Rehabilitation Status Scale (MRSS) of 924 patients with SMIs was 49.82 ± 41.52. The main influencing personal factors of recovery status were age, marital status, education, average annual income and duration of illness. Patients who were accepted by their families tended to have better recovery outcomes, while stigma and social impairment may hinder recovery. Medication adherence played a mediating role between psychosocial factors and recovery status. This study highlights the crucial role of psychosocial factors in the treatment and recovery of SMI patients. Furthermore, medication adherence emerged as a mediating factor. Implementing targeted interventions and policies addressing identified psychosocial barriers can offer promising avenues for improving recovery outcomes for SMI patients in community settings.

Graphene Oxide/Black Phosphorus Functionalized Collagen Scaffolds with Enhanced Near-Infrared Controlled In Situ Biomineralization for Promoting Infectious Bone Defect Repair through PI3K/Akt Pathway.

Infectious bone defects resulting from surgery, infection, or trauma are a prevalent clinical issue. Current treatments commonly used include systemic antibiotics and autografts or allografts. Nevertheless, therapies come with various disadvantages, including multidrug-resistant bacteria, complications arising from the donor site, and immune rejection, which makes artificial implants desirable. However, artificial implants can fail due to bacterial infections and inadequate bone fusion after implantation. Thus, the development of multifunctional bone substitutes that are biocompatible, antibacterial, osteoconductive, and osteoinductive would be of great clinical importance. This study designs and prepares 2D graphene oxide (GO) and black phosphorus (BP) reinforced porous collagen (Col) scaffolds as a viable strategy for treating infectious bone defects. The fabricated Col-GO@BP scaffold exhibited an efficient photothermal antibacterial effect under near-infrared (NIR) irradiation. A further benefit of the NIR-controlled degradation of BP was to promote biomineralization by phosphorus-driven and calcium-extracted phosphorus in situ. The abundant functional groups in GO could synergistically capture the ions and enhance the in situ biomineralization. The Col-GO@BP scaffold facilitated osteogenic differentiation of bone marrow-derived mesenchymal stem cells (BMSC) by leveraging its mild photothermal effect and biomineralization process, which upregulated heat shock proteins (HSPs) and activated PI3K/Akt pathways. Additionally, systematic in vivo experiments demonstrated that the Col-GO@BP scaffold obviously promotes infectious bone repair through admirable photothermal antibacterial performance and enhanced vascularization. As a result of this study, we provide new insights into the photothermal activity of GO@BP nanosheets, their degradation, and a new biological application for them.

The congruence between the preferred and actual places of death among terminal cancer patients in China.

Background: Dying in a preferred place is important for a good death. Currently, no study has evaluated the extent to which the preferences for the place of death (PoD) are met among terminal cancer patients in China. This study examined the congruence between the preferred and actual PoD and its predictors among terminal cancer patients in China. Methods: Between 2015 and 2023, 845 terminal cancer patients from four tertiary hospitals in Wuhan, China, were enrolled and followed till death. Face-to-face surveys at baseline and telephone-based interviews in the last month of patients' lives were combined to learn patients' preferred PoD. Data on patients' actual PoD were collected from families within 1 month after patients' death. Results: Of the 410 patients who died, 62.7% of them died in hospitals. The agreement between patients' preferred and actual PoD was fair (κ = 0.221). The congruence between patients' preferred and actual PoD was 63.0, 36.6%, and 0 for hospital death, home death, and hospice facility/nursing home death separately. Patients were more likely to die in their preferred places if their preferred place and family caregivers' belief of patients' preferred PoD was congruent (odds ratio [OR] = 6.464, p = 0.001), or if caregivers had a medically related occupation (OR = 4.577, p = 0.003); if patients were hospitalized at least twice in the last month of life (OR = 0.422, p = 0.000), or the quality of care received by patients in the last 48 h was rated good by the families (OR = 0.373, p = 0.011), patients were less likely to die in their preferred places. Conclusion: The congruence between patients' preferred and actual PoD was fair. Advance care planning (ACP) needs to be popularized in China, and the quality of care in hospice facilities and nursing homes should be improved. The necessary policy support for hospice care should be made to respect cancer patients' end-of-life (EoL) care preferences in China.

Mutating a flexible region of the RSV F protein can stabilize the prefusion conformation.

The respiratory syncytial virus (RSV) fusion (F) glycoprotein is highly immunogenic in its prefusion (pre-F) conformation. However, the protein is unstable, and its conformation must be stabilized for it to function effectively as an immunogen in vaccines. We present a mutagenesis strategy to arrest the RSV F protein in its pre-F state by blocking localized changes in protein structure that accompany large-scale conformational rearrangements. We generated a series of mutants and screened them in vitro to assess their potential for forming a stable pre-F. In animals, the immunogenicity of a representative mutant F protein, with a conformation confirmed by cryo-electron microscopy, elicited levels of neutralizing antibodies and protection against RSV-induced lung damage that were comparable to those of DS-Cav1, a pre-F used in a licensed vaccine.

The clinical value of ultrasound-guided sacral anesthesia in Intracavitary and/or interstitial brachytherapy for cervical Cancer.

OBJECTIVE: To explore the effectiveness and safety of ultrasound-guided sacral anesthesia in intracavitary and/or interstitial brachytherapy for cervical cancer patients. METHOD: A retrospective analysis was conducted on a total of 1039 intracavitary and/or interstitial brachytherapy involving 220 patients in our department from December 7, 2020 to March 21, 2024. The study assessed the satisfaction with anesthesia, changes in vital signs, onset time of anesthesia, dosage of anesthetic drugs, duration of anesthesia, and incidence of adverse reactions. RESULT: The rate of satisfaction with anesthesia was 73.8 %. There were no significant differences in patients' heart rate (HR), mean arterial pressure (MAP), systolic blood pressure (SBP), or oxygen saturation (SaO2) before and after anesthesia (P > 0.05). The onset time of anesthesia ranged from 20 to 35 min, the average dosage of anesthetic was 20 ml, and the duration of anesthesia lasted from 30 to 120 min. Serious adverse reactions included 2 cases of post-anesthesia syncope (associated with sacral cysts). The incidence of nausea was 2.2 %, and the total incidence of other adverse reactions, such as vomiting, urinary retention, bradycardia, anal discomfort, and dizziness, was less than 1 %. CONCLUSION: Ultrasound guided sacral anesthesia demonstrates significant advantages, including effective anesthesia, minimal impact on vital signs, rapid onset, prolonged maintenance, and a low incidence of adverse reactions. It is recommended for widespread application in intracavitary and/or interstitial brachytherapy for cervical cancer patients.

Monocyte-to-Albumin Ratio Is Associated With Hematoma Expansion in Spontaneous Intracerebral Hemorrhage.

BACKGROUND: Hematoma expansion (HE) after spontaneous intracerebral hemorrhage (ICH) is a severe complication that independently predicts poor prognosis. In this study, we aimed to investigate whether monocyte-to-albumin ratio (MAR), a novel marker of systemic inflammation, could predict HE in patients with ICH. METHODS: We retrospectively assessed the data of patients with ICH. The clinical, imaging, and laboratory test data including, the MAR on admission, were analyzed. A multivariate logistic regression analysis was carried out to explore the relationship between MAR and hematoma growth. The receiver operating characteristic (ROC) curve was employed to investigate the predictive value of MAR for HE after ICH. RESULTS: A total of 246 patients were included in the present study. Multivariate logistic regression analysis demonstrated that the MAR was associated with HE (odds ratio [OR] = 1.179; 95% confidence interval, 1.093-1.272; p = 0.000). ROC curve analysis showed that MAR could predict HE, with an area under the curve of 0.802 (95% CI: 0.744-0.859, p < 0.001). The optimal predictive cutoff value of MAR for HE was 10.01 (sensitivity: 72.43%, specificity: 77.05%). CONCLUSIONS: Our results suggested that a high MAR on admission was associated with an increased risk of HE in ICH patients, and MAR can become an independent predictor of HE in ICH patients.

A predictive model for L-T4 dose in postoperative DTC after RAI therapy and its clinical validation in two institutions.

Purpose: To develop a predictive model using machine learning for levothyroxine (L-T4) dose selection in patients with differentiated thyroid cancer (DTC) after resection and radioactive iodine (RAI) therapy and to prospectively validate the accuracy of the model in two institutions. Methods: A total of 266 DTC patients who received RAI therapy after thyroidectomy and achieved target thyroid stimulating hormone (TSH) level were included in this retrospective study. Sixteen clinical and biochemical characteristics that could potentially influence the L-T4 dose were collected; Significant features correlated with L-T4 dose were selected using machine learning random forest method, and a total of eight regression models were established to assess their performance in prediction of L-T4 dose after RAI therapy; The optimal model was validated through a two-center prospective study (n=263). Results: Six significant clinical and biochemical features were selected, including body surface area (BSA), weight, hemoglobin (HB), height, body mass index (BMI), and age. Cross-validation showed that the support vector regression (SVR) model was with the highest accuracy (53.4%) for prediction of L-T4 dose among the established eight models. In the two-center prospective validation study, a total of 263 patients were included. The TSH targeting rate based on constructed SVR model were dramatically higher than that based on empirical administration (Rate 1 (first rate): 52.09% (137/263) vs 10.53% (28/266); Rate 2 (cumulative rate): 85.55% (225/263) vs 53.38% (142/266)). Furthermore, the model significantly shortens the time (days) to achieve target TSH level (62.61 ± 58.78 vs 115.50 ± 71.40). Conclusions: The constructed SVR model can effectively predict the L-T4 dose for postoperative DTC after RAI therapy, thus shortening the time to achieve TSH target level and improving the quality of life for DTC patients.

A Nomogram of Predicting Healthcare-Associated Infections in Burned Children.

BACKGROUND: Healthcare-associated infections (HAIs) are a common clinical concern associated with adverse prognosis and mortality in burned children. This study aimed to construct a predictive nomogram of the risk of HAIs in burned children. METHODS: Children admitted to the burn unit of Wuhan Third Hospital between 2020 and 2022 were included. The univariate and multivariate logistic regression analyses were adopted to ascertain predictors of HAIs. A nomogram was developed to predict the HAI risk of each patient, with receiver operating characteristic curves and calibration curves being generated to assess its predictive ability. Furthermore, decision and impact curves were used to assess the clinical utility. RESULTS: Of 1122 burned children, 61 (5.5%) patients experienced HAIs. The multivariate analysis indicated that total burn surface area, length of stay, surgery, central venous catheter use and urinary catheter use were the independent risk factors of HAIs. Using these variables, we developed a predictive nomogram of the occurrence of HAIs in burned children, and the internal validation results demonstrated good discrimination and calibration of the nomogram. The area under the curve values of the nomogram was 0.926 (95% CI, 0.896-0.957). The calibration curve showed high consistency between the actual and predicted HAIs. The decision and impact curve indicated that the nomogram was of good clinical utility and more credible net clinical benefits in predicting HAIs. CONCLUSIONS: The present study constructed a nomogram for predicting the risk of HAIs in burned children. This nomogram may strengthen the effective screening of patients at high risk of HAIs.

Comprehensive analysis of benzothiazoles (BTHs), benzotriazoles (BTRs), and benzotriazole ultraviolet absorbers (BUVs) in the western South China Sea: Spatial distributions, migration tendencies and ecotoxicological relevance.

Benzothiazoles (BTHs), benzotriazoles (BTRs), and benzotriazole ultraviolet absorbers (BUVs) have garnered significant attention owing to their persistent nature in the environment and adverse impacts on aquatic organisms. However, there remains a dearth of investigations and studies conducted in tropical marine environments. In this study, we undertook the inaugural distributional survey and ecotoxicological relevance of BTHs, BTRs, and BUVs in seawater and sediments of the western South China Sea (WSCS). Elevated concentrations of BTHs, BTRs, and BUVs in the seawater and suspended particulate matter (SPM) were primarily observed in the Pearl River Estuary (PRE) and the western region of the WSCS, attributed to terrestrial runoff and hydrodynamic processes. Moreover, the transport of these compounds at the seawater-SPM interface was influenced by both the intrinsic properties of the contaminants and temperature variations. Spatially, concentrations of BTHs, BTRs, and BUVs in surface sediments exhibited a diminishing trend with increasing distance from the coast to offshore areas, reflecting notable anthropogenic impacts. Concentration profiles of these compounds in sediment cores displayed a bottom-up increasing trend, with total organic carbon (TOC) identified as the primary determinant governing their accumulation within sediment cores in the WSCS. Terrestrial runoff inputs and atmospheric deposition as major contributors to the occurrence of BTHs, BTRs, and BUVs in the WSCS. Simultaneously, the study underscores the non-negligible moderate mixture risk quotient associated with BTHs, BTRs, and BUVs in the sediments.

A survival nomogram involving nutritional-inflammatory indicators for cervical cancer patients receiving adjuvant radiotherapy.

Objective: The combined impact of nutritional and inflammatory status on survival of cervical cancer patients remained unclear. This study aimed to construct a survival nomogram involving both nutritional and inflammatory indicators and evaluate their potential correlation. Methods: This retrospective study included 325 cervical cancer patients who received adjuvant radiotherapy between September 2010 and September 2020. Baseline nutritional indicators such as body mass index (BMI), controlling nutritional status (CONUT) and serum albumin were assessed. Inflammatory indicators of platelet/lymphocyte ratio (PLR), neutrophil/lymphocyte ratio (NLR), systemic immune inflammation index (SII) and system inflammation response index (SIRI) were evaluated respectively. The LASSO regression and Cox regression models were applied for variable selection and nomogram building. The predictive accuracy and superiority of prognostic model were assessed by area under curve (AUC), C-index, decision curve analysis (DCA), integrated discrimination improvement (IDI) and net reclassification improvement (NRI). Results: Patients with high inflammatory indicators (PLR, NLR and SII) and poor nutritional status (CONUT scores > 2) suffered poorer prognosis compared to these with well nutritional status and lower inflammation levels. Our study unveiled a positive correlation between malnutrition and hyperinflammation. Even after accounting for baseline inflammatory level, malnutrition remained a significant risk factor for patients. Notably, the inflammatory level and nutritional status were further modulated by the clinical features of patients. Patients with poorer nutritional status exhibited higher levels of PLR, NLR, SII and SIRI, particularly for those in advanced clinical stages and with non-squamous cell carcinoma. In addition, our study found elevated levels of circulating basophil and serum carbohydrate antigen 125 (CA125) were associated with the poor prognosis. The prognostic nomogram which incorporated the nutritional-inflammatory indicators of PLR and CONUT showed a favorable performance with the AUC value of 0.76 at 5-year survival prediction. The DCA, IDI and NRI consistently demonstrated the favorable superiority of the model. Moreover, the nomogram-based risk stratification system could effectively classify patients into three mortality risks subgroups. Conclusions: Poorer nutritional and high inflammatory status collectively contributed to the poorer prognosis. The prognostic nomogram which incorporated nutritional-inflammatory indicators significantly improved the prediction of long-term outcomes of cervical cancer patients undergoing adjuvant radiotherapy.

Identification and characterization of new Siberian subtype of tick-borne encephalitis virus isolates revealed genetic variations of the Chinese strains.

Tick-borne encephalitis virus (TBEV) is a pathogen that causes febrile infectious diseases and neurological damage to humans. TBEVs are prevalent from Europe to Far Eastern Asia, including Northeastern China. The understanding of TBEV phylogeny in China has been limited owing to insufficient genomic data on Chinese TBEV strains. Here, six TBEV strains were isolated from ticks collected in Inner Mongolia. The transmission electron microscopy revealed spherical particles with an enveloped structure of 50-60 nm in diameter. Phylogenetic analysis showed that, two strains were classified as the Siberian subtype, while the remaining four were identified as the Far Eastern subtype. Migration analyses based on TBEV ORF and envelope (E) protein sequences revealed that Chinese TBEV strains were migrated from Russia and/or Kazakhstan into China. Hulun Buir and Mudanjiang, the northeastern region of China, are considered hotspots with multiple import and export routes of Chinese TBEV strains. These results promote the understanding of TBEV genetic variations and phylogeny in China and suggest the importance of improving investigation of TBEV prevalence, which would instrumental for vaccine design strategies and better preparation for controlling TBEV infection in humans.

Size-resolved gas-particle partitioning of polycyclic aromatic hydrocarbons in a large temperature range of 50 ℃.

Size-resolved gas-particle partitioning of semi-volatile organic compounds (SVOCs) can affect their environmental behaviors and health effects, which has not been widely studied in comparing with the gas-total suspended particle partitioning. Herein, the size-resolved gas-particle partitioning quotient (KPi) of polycyclic aromatic hydrocarbons (PAHs) in a large temperature range (-20.6 â„ƒ ∼ 29.4 â„ƒ) was firstly comprehensively studied. The log KPi values of PAHs related to fine particles were significantly higher than those related to coarse particles. When the logarithm of subcooled liquid-vapor pressure (log PL0) ∈ [-7, -1), the regression slopes of log KPi vs log PL0 related to the particle size > 1.0 µm were shallower than those with the particle size range of 0.10-1.0 µm, which indicated the influence of particle size on KPi. Among the three previous prediction equations of gas-particle partitioning quotient, the empirical equation based on the ambient temperature matched better with the measured log KPi. Therefore, a new prediction equation including ambient temperature and particle size as the two major parameters was established. For most particle size ranges, the new equation showed better prediction performance than the three previous equations. In summary, this study provided new insights for the size-resolved gas-particle partitioning mechanism and quotient.

Catalytic divergence of O-methyltransferases shapes the chemo-diversity of polymethoxylated bibenzyls in Dendrobium catenatum.

Erianin, crepidatin, and chrysotobibenzyl are typical medicinal polymethoxylated bibenzyls (PMBs) that are commercially produced in Dendrobium species. PMBs' chemo-diversity is mediated by the manifold combinations of O-methylation and hydroxylation in a definite order, which remains unsolved. To unequivocally elucidate the methylation mechanism of PMBs, 15 possible intermediates in the biosynthetic pathway of PMBs were chemically synthesized. DcOMT1-5 were highly expressed in tissues where PMBs were biosynthesized, and their expression patterns were well-correlated with the accumulation profiles of PMBs. Moreover, cell-free orthogonal tests based on the synthesized intermediates further confirmed that DcOMT1-5 exhibited distinct substrate preferences and displayed hydroxyl-group regiospecificity during the sequential methylation process. The stepwise methylation of PMBs was discovered from SAM to dihydro-piceatannol (P) in the following order: P → 3-MeP → 4-OH-3-MeP → 4-OH-3,5-diMeP → 3,3'(4'),5-triMeP → 3,4,4',5-tetraMeP (erianin) or 3,3',4,5-tetraMeP (crepidatin) → 3,3',4,4',5-pentaMeP (chrysotobibenzyl). Furthermore, the regioselectivities of DcOMTs were investigated by ligand docking analyses which corresponded precisely with the catalytic activities. In summary, the findings shed light on the sequential catalytic mechanisms of PMB biosynthesis and provide a comprehensive PMB biosynthetic network in D. catenatum. The knowledge gained from this study may also contribute to the development of plant-based medicinal applications and the production of high-value PMBs.

Radiomics based on 18F-FDG PET/CT for prediction of pathological complete response to neoadjuvant therapy in non-small cell lung cancer.

Purpose: This study aimed to establish and evaluate the value of integrated models involving 18F-FDG PET/CT-based radiomics and clinicopathological information in the prediction of pathological complete response (pCR) to neoadjuvant therapy (NAT) for non-small cell lung cancer (NSCLC). Methods: A total of 106 eligible NSCLC patients were included in the study. After volume of interest (VOI) segmentation, 2,016 PET-based and 2,016 CT-based radiomic features were extracted. To select an optimal machine learning model, a total of 25 models were constructed based on five sets of machine learning classifiers combined with five sets of predictive feature resources, including PET-based alone radiomics, CT-based alone radiomics, PET/CT-based radiomics, clinicopathological features, and PET/CT-based radiomics integrated with clinicopathological features. Area under the curves (AUCs) of receiver operator characteristic (ROC) curves were used as the main outcome to assess the model performance. Results: The hybrid PET/CT-derived radiomic model outperformed PET-alone and CT-alone radiomic models in the prediction of pCR to NAT. Moreover, addition of clinicopathological information further enhanced the predictive performance of PET/CT-derived radiomic model. Ultimately, the support vector machine (SVM)-based PET/CT radiomics combined clinicopathological information presented an optimal predictive efficacy with an AUC of 0.925 (95% CI 0.869-0.981) in the training cohort and an AUC of 0.863 (95% CI 0.740-0.985) in the test cohort. The developed nomogram involving radiomics and pathological type was suggested as a convenient tool to enable clinical application. Conclusions: The 18F-FDG PET/CT-based SVM radiomics integrated with clinicopathological information was an optimal model to non-invasively predict pCR to NAC for NSCLC.

Prediction of mutation-induced protein stability changes based on the geometric representations learned by a self-supervised method.

BACKGROUND: Thermostability is a fundamental property of proteins to maintain their biological functions. Predicting protein stability changes upon mutation is important for our understanding protein structure-function relationship, and is also of great interest in protein engineering and pharmaceutical design. RESULTS: Here we present mutDDG-SSM, a deep learning-based framework that uses the geometric representations encoded in protein structure to predict the mutation-induced protein stability changes. mutDDG-SSM consists of two parts: a graph attention network-based protein structural feature extractor that is trained with a self-supervised learning scheme using large-scale high-resolution protein structures, and an eXtreme Gradient Boosting model-based stability change predictor with an advantage of alleviating overfitting problem. The performance of mutDDG-SSM was tested on several widely-used independent datasets. Then, myoglobin and p53 were used as case studies to illustrate the effectiveness of the model in predicting protein stability changes upon mutations. Our results show that mutDDG-SSM achieved high performance in estimating the effects of mutations on protein stability. In addition, mutDDG-SSM exhibited good unbiasedness, where the prediction accuracy on the inverse mutations is as well as that on the direct mutations. CONCLUSION: Meaningful features can be extracted from our pre-trained model to build downstream tasks and our model may serve as a valuable tool for protein engineering and drug design.