MetDIT: Transforming and Analyzing Clinical Metabolomics Data with Convolutional Neural Networks.

Clinical metabolomics is growing as an essential tool for precision medicine. However, classical machine learning algorithms struggle to comprehensively encode and analyze the metabolomics data due to their high dimensionality and complex intercorrelations. This article introduces a new method called MetDIT, designed to analyze intricate metabolomics data effectively using deep convolutional neural networks (CNN). MetDIT comprises two components: TransOmics and NetOmics. Since CNN models have difficulty in processing one-dimensional (1D) sequence data efficiently, we developed TransOmics, a framework that transforms sequence data into two-dimensional (2D) images while maintaining a one-to-one correspondence between the sequences and images. NetOmics, the second component, leverages a CNN architecture to extract more discriminative representations from the transformed samples. To overcome the overfitting due to the small sample size and class imbalance, we introduced a feature augmentation module (FAM) and a loss function to improve the model performance. Furthermore, we systematically optimized the model backbone and image resolution to balance the model parameters and computational costs. To demonstrate the performance of the proposed MetDIT, we conducted extensive experiments using three different clinical metabolomics data sets and achieved better classification performance than classical machine learning methods used in metabolomics, including Random Forest, SVM, XGBoost, and LightGBM. The source code is available at the GitHub repository at https://github.com/Li-OmicsLab/MetDIT, and the WebApp can be found at http://metdit.bioinformatics.vip/.

Metabolic Connectome and Its Role in the Prediction, Diagnosis, and Treatment of Complex Diseases.

The interconnectivity of advanced biological systems is essential for their proper functioning. In modern connectomics, biological entities such as proteins, genes, RNA, DNA, and metabolites are often represented as nodes, while the physical, biochemical, or functional interactions between them are represented as edges. Among these entities, metabolites are particularly significant as they exhibit a closer relationship to an organism's phenotype compared to genes or proteins. Moreover, the metabolome has the ability to amplify small proteomic and transcriptomic changes, even those from minor genomic changes. Metabolic networks, which consist of complex systems comprising hundreds of metabolites and their interactions, play a critical role in biological research by mediating energy conversion and chemical reactions within cells. This review provides an introduction to common metabolic network models and their construction methods. It also explores the diverse applications of metabolic networks in elucidating disease mechanisms, predicting and diagnosing diseases, and facilitating drug development. Additionally, it discusses potential future directions for research in metabolic networks. Ultimately, this review serves as a valuable reference for researchers interested in metabolic network modeling, analysis, and their applications.

[Evaluation of extravascular lung water index in critically ill patients based on lung ultrasound radiomics analysis combined with machine learning].

OBJECTIVE: To explore lung ultrasound radiomics features which related to extravascular lung water index (EVLWI), and to predict EVLWI in critically ill patients based on lung ultrasound radiomics combined with machine learning and validate its effectiveness. METHODS: A retrospective case-control study was conducted. The lung ultrasound videos and pulse indicated continuous cardiac output (PiCCO) monitoring results of critically ill patients admitted to the department of critical care medicine of the First Affiliated Hospital of Guangxi Medical University from November 2021 to October 2022 were collected, and randomly divided into training set and validation set at 8:2. The corresponding images from lung ultrasound videos were obtained to extract radiomics features. The EVLWI measured by PiCCO was regarded as the "gold standard", and the radiomics features of training set was filtered through statistical analysis and LASSO algorithm. Eight machine learning models were trained using filtered radiomics features including random forest (RF), extreme gradient boost (XGBoost), decision tree (DT), Naive Bayes (NB), multi-layer perceptron (MLP), K-nearest neighbor (KNN), support vector machine (SVM), and Logistic regression (LR). Receiver operator characteristic curve (ROC curve) was plotted to evaluate the predictive performance of models on EVLWI in the validation set. RESULTS: A total of 151 samples from 30 patients were enrolled (including 906 lung ultrasound videos and 151 PiCCO monitoring results), 120 in the training set, and 31 in the validation set. There were no statistically significant differences in main baseline data including gender, age, body mass index (BMI), mean arterial pressure (MAP), central venous pressure (CVP), heart rate (HR), cardiac index (CI), cardiac function index (CFI), stroke volume index (SVI), global end diastolic volume index (GEDVI), systemic vascular resistance index (SVRI), pulmonary vascular permeability index (PVPI) and EVLWI. The overall EVLWI range in 151 PiCCO monitoring results was 3.7-25.6 mL/kg. Layered analysis showed that both datasets had EVLWI in the 7-15 mL/kg interval, and there was no statistically significant difference in EVLWI distribution. Two radiomics features were selected by using LASSO algorithm, namely grayscale non-uniformity (weight was -0.006 464) and complexity (weight was -0.167 583), and they were used for modeling. ROC curve analysis showed that the MLP model had better predictive performance. The area under the ROC curve (AUC) of the prediction validation set EVLWI was higher than that of RF, XGBoost, DT, KNN, LR, SVM, NB models (0.682 vs. 0.658, 0.657, 0.614, 0.608, 0.596, 0.557, 0.472). CONCLUSIONS: The gray level non-uniformity and complexity of lung ultrasound were the most correlated radiomics features with EVLWI monitored by PiCCO. The MLP model based on gray level non-uniformity and complexity of lung ultrasound can be used for semi-quantitative prediction of EVLWI in critically ill patients.

[Research progress on transcranial electrical stimulation for deep brain stimulation].

Transcranial electric stimulation (TES) is a non-invasive, economical, and well-tolerated neuromodulation technique. However, traditional TES is a whole-brain stimulation with a small current, which cannot satisfy the need for effectively focused stimulation of deep brain areas in clinical treatment. With the deepening of the clinical application of TES, researchers have constantly investigated new methods for deeper, more intense, and more focused stimulation, especially multi-electrode stimulation represented by high-precision TES and temporal interference stimulation. This paper reviews the stimulation optimization schemes of TES in recent years and further analyzes the characteristics and limitations of existing stimulation methods, aiming to provide a reference for related clinical applications and guide the following research on TES. In addition, this paper proposes the viewpoint of the development direction of TES, especially the direction of optimizing TES for deep brain stimulation, aiming to provide new ideas for subsequent research and application.

Novel clinical biomarkers in blood and pleural effusion for diagnosing patients with tuberculosis distinguishing from malignant tumor.

Pleural effusion (PE) is a common manifestation of tuberculosis (TB) and malignant tumors but tuberculous PE (TPE) is difficult to distinguish from malignant PE (MPE), especially by noninvasive detection indicators. This study aimed to find effective detection indices in blood and PE for differentiating TB from a malignant tumor. A total of 815 patients who were diagnosed with TB or cancer in Hubei Shiyan Taihe Hospital from 2014 to 2017 were collected. Amongst them, 717 were found to have PE by thoracoscopy. Clinical characteristics, patients' blood parameters and PE indicator information were summarized for analysis. Patients with MPE had higher percentages to be bloody and negative of Rivalta test in PE than those with TPE. For clinical indicators, comparison of the specific parameters in blood showed that 18 indicators were higher in the TPE group than in the MPE group. By contrast, 12 indicators were higher in the MPE group than in the TPE group (P < .01). In addition, in PE tests, 3 parameters were higher in the TPE group, whereas other 4 parameters were higher in the MPE group (P < .01). Then, for clinical diagnosing practice, ROC analysis and principal component analysis were applied. The top 6 relevant indicators with area under curve over 0.70 were screened out as follows: hydrothorax adenosine dehydrogenase (pADA, 0.90), hydrothorax high-sensitivity C reactive protein (0.79), percentage of blood monocyte (sMONp, 0.75), blood high-sensitivity C reactive protein (sHsCRP, 0.73), erythrocyte sedimentation rate (0.71) and blood D-dimer (0.70). Moreover, logistic regression model revealed that a specific combination of 3 biomarkers, namely, pADA, sMONp and sHsCRP, could enhance the distinguishment of TB from malignant tumor with PE (area under curve = 0.944, 95% confidence interval = 0.925-0.964). The diagnostic function of the top single marker pADA in patients from different groups was analyzed and it was found to maintain high specificity and sensitivity. The 6 indicators, namely, pADA, hydrothorax high-sensitivity C reactive protein, sMONp, sHsCRP, sESR and blood D-dimer, showed significant diagnostic value for clinicians. Further, the combination of pADA, sMONp and sHsCRP has high accuracy for differential diagnosis for the first time. Most interestingly, the single marker pADA maintained high specificity and sensitivity in patients with different statuses and thus has great value for rapid and accurate diagnosis of suspected cases.

The key role of sphingolipid metabolism in cancer: New therapeutic targets, diagnostic and prognostic values, and anti-tumor immunotherapy resistance.

Biologically active sphingolipids are closely related to the growth, differentiation, aging, and apoptosis of cancer cells. Some sphingolipids, such as ceramides, are favorable metabolites in the sphingolipid metabolic pathway, usually mediating antiproliferative responses, through inhibiting cancer cell growth and migration, as well as inducing autophagy and apoptosis. However, other sphingolipids, such as S1P, play the opposite role, which induces cancer cell transformation, migration and growth and promotes drug resistance. There are also other sphingolipids, as well as enzymes, played potentially critical roles in cancer physiology and therapeutics. This review aimed to explore the important roles of sphingolipid metabolism in cancer. In this article, we summarized the role and value of sphingolipid metabolism in cancer, including the distribution of sphingolipids, the functions, and their relevance to cancer diagnosis and prognosis. We also summarized the known and potential antitumor targets present in sphingolipid metabolism, analyzed the correlation between sphingolipid metabolism and tumor immunity, and summarize the antitumor effects of natural compounds based on sphingolipids. Through the analysis and summary of sphingolipid antitumor therapeutic targets and immune correlation, we aim to provide ideas for the development of new antitumor drugs, exploration of new therapeutic means for tumors, and study of immunotherapy resistance mechanisms.

Andrographolide suppresses non-small-cell lung cancer progression through induction of autophagy and antitumor immune response.

Despite recent advances in diagnosis and therapeutic strategies, treatment of non-small-cell lung cancer (NSCLC) remains unsatisfactory in terms of prognosis. Andrographolide (AD), a principal active component of Andrographis paniculata (Burm.f.) Nees, exerts anti-cancer therapeutic properties. AD has been used for centuries in China for clinical treatment of viral infections. However, the pharmacological biology of AD in NSCLC remains unknown. In this study, AD regulated autophagy and PD-L1 expression in NSCLC. Molecular dynamics simulations indicated that AD bound directly to signal transducer and activator of transcription-3 (STAT3) with high affinity. Proteomics analysis indicated that AD reduced the expression of tumour PD-L1 in NSCLC by suppressing JAK2/STAT3 signalling. AD modulated the P62-dependent selective autophagic degradation of PD-L1 by inhibiting STAT3 phosphorylation. In vivo study revealed that AD suppressed tumour growth in H1975 xenograft mice and Lewis lung carcinoma cell models, and better efficacy was obtained at higher concentrations. AD prolonged the survival time of the mice and enhanced the treatment efficacy of anti-PD-1 mAb immunotherapy by stimulating CD8+ T cell infiltration and function. This work elucidated the specific mechanism by which AD inhibited NSCLC. Treatment with the combination of AD and anti-PD-1 mAb immunotherapy could be a potential strategy for patients with NSCLC.

Clinico-Radiologic Nomogram Using Multiphase CT to Predict Lymph Node Metastasis in Colon Cancer.

PURPOSE: To evaluate the value of multiphase computed tomography (CT)-based radiomics for predicting lymph node metastasis in patients with colorectal cancer (CRC). METHODS: This study included 191 patients enrolled in our hospital who underwent non-contrast, arterial, and portal venous phase CT scans between June 2017 and December 2019. Segmented regions of interest in each slice of CT images were used to extract radiomics features. Redundant features were ruled out using the least absolute shrinkage and selection operator (LASSO) regression. The multiphase CT-combined radiomics signature (Com-RS) was constructed based on the selected radiomics features from the three CT phases weighted by the respective LASSO coefficients. The nomogram was created by combining the Com-RS with key clinical parameters. The performance of the nomogram was evaluated using receiver operating characteristics, calibration, and decision curve analyses (DCA). RESULTS: Nine features were demonstrated to be the most significant and used to build the Com-RS: two from non-contrast CT, four from arterial CT, and three from portal venous CT. Tumor length has been identified as a key clinical parameter. A radiomics nomogram was constructed by integrating the Com-RS with tumor length and generated good performance with areas under the curve of 0.830 (95% confidence interval [CI], 0.758 - 0.902) and 0.712 (95% CI, 0.585 - 0.839) in the training and validation cohorts, respectively. Calibration and DCA confirmed the potential clinical relevance and applicability of the nomogram. CONCLUSIONS: The developed multiphase CT-based radiomics nomogram can potentially serve as an effective tool for the preoperative prediction of lymph node status in CRC.

Emodin induces apoptosis and suppresses non-small-cell lung cancer growth via downregulation of sPLA2-IIa.

BACKGROUND: Lung cancer has become the principal cause of cancer-related deaths. Emodin is a Chinese herb-derived compound extracted from the roots of Rheum officinale that exhibits numerous pharmacological characteristics. Secretory phospholipase A2-IIa (sPLA2-IIa) is overexpressed in cancers and plays an important role in cancer development. PURPOSE: This study aims to investigate the anti-tumor mechanism of emodin in non-small-cell lung cancer (NSCLC). METHODS: MTT assay was applied to detect the sensitivity of emodin to NSCLC cell line. Flow cytometry was used to examine the effect of emodin on cell cycle distribution and evaluate ROS level and apoptosis. Western blot analysis was utilised to examine the expression levels of sPLA2-IIa, PKM2, and AMPK and its downstream pathways induced by emodin. Enzyme inhibition assay was applied to investigate the inhibitory effect of emodin on sPLA2-IIa. The anticancer effect of emodin was also detected using an in vivo model. RESULTS: Emodin significantly inhibited NSCLC proliferation in vivo and in vitro and was relatively less cytotoxic to normal lung cell lines. Most importantly, emodin inhibited the proliferation of KRAS mutant cell lines by decreasing the expression of sPLA2-IIa and NF-κB pathways. Emodin also inhibited mTOR and AKT and activated the AMPK pathway. Furthermore, emodin induced apoptosis, increased the reactive oxygen species (ROS) level, and arrested the cell cycle. CONCLUSION: Emodin exhibited a novel anti-tumor mechanism of inhibiting the proliferation of KRAS mutant cell lines by decreasing the expression levels of sPLA2-IIa and NF-κB pathways. Hence, emodin can potentially serve as a therapeutic target in NSCLC.

Potential prognostic factors in progression-free survival for patients with cervical cancer.

BACKGROUND: Cervical cancer continues to be one of the leading causes of cancer deaths among females in low and middle-income countries. In this study, we aimed to assess the independent prognostic value of clinical and potential prognostic factors in progression-free survival (PFS) in cervical cancer. METHODS: We conducted a retrospective study on 92 cervical cancer patients treated from 2017 to 2019 at the Zhuhai Hospital of Traditional Chinese and Western Medicine. Tumor characteristics, treatment options, progression-free survival and follow-up information were collected. Kaplan-Meier method was used to assess the PFS. RESULTS: Results showed that the number of retrieved lymph nodes had a statistically significant effect on PFS of cervical cancer patients (P = 0.002). Kaplan-Meier survival curve analysis showed that cervical cancer patients with initial symptoms age 25-39 had worse survival prognoses (P = 0.020). And the using of uterine manipulator in laparoscopic treatment showed a better prognosis (P < 0.001). A novel discovery of our study was to verify the prognostic values of retrieved lymph nodes count combining with FIGO staging system, which had never been investigated in cervical cancer before. According to the Kaplan-Meier survival curve analysis and receiver operating characteristic (ROC) curve analysis, significant improvements were found after the combination of retrieved lymph nodes count and FIGO stage in predicting PFS for cervical cancer patients (P < 0.001, AUC = 0.826, 95% CI: 0.689-0.962). CONCLUSION: Number of retrieved lymph nodes, initial symptoms age, uterine manipulator, and retrieved lymph nodes count combining with FIGO staging system could be potential prognostic factors for cervical cancer patients.

Early lung cancer diagnostic biomarker discovery by machine learning methods.

Early diagnosis has been proved to improve survival rate of lung cancer patients. The availability of blood-based screening could increase early lung cancer patient uptake. Our present study attempted to discover Chinese patients' plasma metabolites as diagnostic biomarkers for lung cancer. In this work, we use a pioneering interdisciplinary mechanism, which is firstly applied to lung cancer, to detect early lung cancer diagnostic biomarkers by combining metabolomics and machine learning methods. We collected total 110 lung cancer patients and 43 healthy individuals in our study. Levels of 61 plasma metabolites were from targeted metabolomic study using LC-MS/MS. A specific combination of six metabolic biomarkers note-worthily enabling the discrimination between stage I lung cancer patients and healthy individuals (AUC = 0.989, Sensitivity = 98.1%, Specificity = 100.0%). And the top 5 relative importance metabolic biomarkers developed by FCBF algorithm also could be potential screening biomarkers for early detection of lung cancer. Naïve Bayes is recommended as an exploitable tool for early lung tumor prediction. This research will provide strong support for the feasibility of blood-based screening, and bring a more accurate, quick and integrated application tool for early lung cancer diagnostic. The proposed interdisciplinary method could be adapted to other cancer beyond lung cancer.