DeepFold: enhancing protein structure prediction through optimized loss functions, improved template features, and re-optimized energy function.

MOTIVATION: Predicting protein structures with high accuracy is a critical challenge for the broad community of life sciences and industry. Despite progress made by deep neural networks like AlphaFold2, there is a need for further improvements in the quality of detailed structures, such as side-chains, along with protein backbone structures. RESULTS: Building upon the successes of AlphaFold2, the modifications we made include changing the losses of side-chain torsion angles and frame aligned point error, adding loss functions for side chain confidence and secondary structure prediction, and replacing template feature generation with a new alignment method based on conditional random fields. We also performed re-optimization by conformational space annealing using a molecular mechanics energy function which integrates the potential energies obtained from distogram and side-chain prediction. In the CASP15 blind test for single protein and domain modeling (109 domains), DeepFold ranked fourth among 132 groups with improvements in the details of the structure in terms of backbone, side-chain, and Molprobity. In terms of protein backbone accuracy, DeepFold achieved a median GDT-TS score of 88.64 compared with 85.88 of AlphaFold2. For TBM-easy/hard targets, DeepFold ranked at the top based on Z-scores for GDT-TS. This shows its practical value to the structural biology community, which demands highly accurate structures. In addition, a thorough analysis of 55 domains from 39 targets with publicly available structures indicates that DeepFold shows superior side-chain accuracy and Molprobity scores among the top-performing groups. AVAILABILITY AND IMPLEMENTATION: DeepFold tools are open-source software available at https://github.com/newtonjoo/deepfold.

Cancer-Associated Fibroblasts Together with a Decline in CD8+ T Cells Predict a Worse Prognosis for Breast Cancer Patients.

BACKGROUND: Cancer-associated fibroblasts (CAFs) play a crucial role in tumor microenvironment regulation and cancer progression. This study assessed the significance and predictive potential of CAFs in breast cancer prognosis. METHODS: The study included 1503 breast cancer patients. Cancer-associated fibroblasts were identified using morphologic features from hematoxylin and eosin slides. The study analyzed clinicopathologic parameters, survival rates, immune cells, gene sets, and prognostic models using gene-set enrichment analysis, in silico cytometry, pathway analysis, in vitro drug-screening, and gradient-boosting machine (GBM)-learning. RESULTS: The presence of CAFs correlated significantly with young age, lymphatic invasion, and perineural invasion. In silico cytometry showed altered leukocyte subsets in the presence of CAFs, with decreased CD8+ T cells. Gene-set enrichment analysis showed associations with critical processes such as the epithelial-mesenchymal transition and immune modulation. Drug sensitivity analysis in breast cancer cell lines with varying fibroblast activation protein-α expression suggested that CAF-targeted therapies might enhance the efficacy of certain anticancer drugs including ARRY-520, ispinesib-mesylate, paclitaxel, and docetaxel. Integrating CAF presence with machine-learning improved survival prediction. For breast cancer patients, CAFs were independent prognostic markers for worse disease-specific survival and disease-free survival. CONCLUSION: This study highlighted the significance of CAFs in breast cancer biology and provided compelling evidence of their impact on patient outcomes and treatment response. The findings offer valuable insights into the potential of CAFs as prognostic and predictive biomarkers and support the development of CAF-targeted therapies to improve breast cancer management.

JAK2 Loss Arising From Tumor-Spread-Through-Air-Spaces (STAS) Promotes Tumor Progression by Suppressing CD8+ T Cells in Lung Adenocarcinoma: A Machine Learning Approach.

BACKGROUND: Tumor spread through air spaces (STAS) is a recently discovered risk factor for lung adenocarcinoma (LUAD). The aim of this study was to investigate specific genetic alterations and anticancer immune responses related to STAS. By using a machine learning algorithm and drug screening in lung cancer cell lines, we analyzed the effect of Janus kinase 2 (JAK2) on the survival of patients with LUAD and possible drug candidates. METHODS: This study included 566 patients with LUAD corresponding to clinicopathological and genetic data. For analyses of LUAD, we applied gene set enrichment analysis (GSEA), in silico cytometry, pathway network analysis, in vitro drug screening, and gradient boosting machine (GBM) analysis. RESULTS: The patients with STAS had a shorter survival time than those without STAS (P < 0.001). We detected gene set-related downregulation of JAK2 associated with STAS using GSEA. Low JAK2 expression was related to poor prognosis and a low CD8+ T-cell fraction. In GBM, JAK2 showed improved survival prediction performance when it was added to other parameters (T stage, N stage, lymphovascular invasion, pleural invasion, tumor size). In drug screening, mirin, CCT007093, dihydroretenone, and ABT737 suppressed the growth of lung cancer cell lines with low JAK2 expression. CONCLUSION: In LUAD, low JAK2 expression linked to the presence of STAS might serve as an unfavorable prognostic factor. A relationship between JAK2 and CD8+ T cells suggests that STAS is indirectly related to the anticancer immune response. These results may contribute to the design of future experimental research and drug development programs for LUAD with STAS.

High DKK3 expression related to immunosuppression was associated with poor prognosis in glioblastoma: machine learning approach.

BACKGROUND: Glioblastoma multiforme (GBM) is an aggressive malignant primary brain tumor. Wnt/ß-catenin is known to be related to GBM stemness. Cancer stem cells induce immunosuppressive and treatment resistance in GBM. We hypothesized that Wnt/ß-catenin-related genes with immunosuppression could be related to the prognosis in patients with GBM. METHODS: We obtained the clinicopathological data of 525 patients with GBM from the brain cancer gene database. The fraction of tumor-infiltrating immune cells was evaluated using in silico flow cytometry. Among gene sets of Wnt/ß-catenin pathway, Dickkopf-3 (DKK3) gene related to the immunosuppressive response was found using machine learning. We performed gene set enrichment analysis (GSEA), network-based analysis, survival analysis and in vitro drug screening assays based on Dickkopf-3 (DKK3) expression. RESULTS: In analyses of 31 genes related to Wnt/ß-catenin signaling, high DKK3 expression was negatively correlated with increased antitumoral immunity, especially CD8 + and CD4 + T cells, in patients with GBM. High DKK3 expression was correlated with poor survival and disease progression in patients with GBM. In pathway-based network analysis, DKK3 was directly linked to the THY1 gene, a tumor suppressor gene. Through in vitro drug screening, we identified navitoclax as an agent with potent activity against GBM cell lines with high DKK3 expression. CONCLUSIONS: These results suggest that high DKK3 expression could be a therapeutic target in GBM. The results of the present study could contribute to the design of future experimental research and drug development programs for GBM.

Sex Difference in Coronary Artery Spasm Tested by Intracoronary Acetylcholine Provocation Test in Patients with Nonobstructive Coronary Artery Disease.

Introduction: Cardiovascular diseases manifest differently in men and women. The purpose of this study is to compare the sex difference in the characteristics of coronary artery spasm (CAS) in patients with nonobstructive cardiovascular disease (NOCVD) and the clinical outcomes in accordance with sex in CAS patients. Methods: The study analysed 5,491 patients with NOCVD who underwent an acetylcholine provocation test from November 2004 to May 2014 for evaluation of chest pain. CAS was defined as greater than 70% of luminal narrowing of the artery during the acetylcholine provocation test. Results: The patients were divided into men (n = 2,506) and women (n = 2,985). Mean follow-up days were 1,218 ± 577 days. To adjust for confounding factors, the propensity score matching (PSM) analysis was performed in all patients and among the CAS patients. After PSM analysis, a total of 1,201 pairs in all patients and a total of 713 pairs in CAS patients were generated. In all patients, women showed significantly less incidence of CAS compared with men (62.3% vs 50.9%, P < 0.01). Myocardial bridge (MB) and moderate stenosis were less prevalent in women, while transient ST elevation and ischemic chest pain during provocation were more frequent in women. In CAS patients, men had a higher incidence of multivessel spasm than women (35.7% vs. 29.7%, P < 0.01). Old age, dyslipidemia, and MB were independent risk factors of CAS in both men and women. In CAS patients, there was no statistical differences for various individual and composite major outcomes up to five years in either men or women. In men with CAS, old age was a risk factor of a 5-year major adverse cardiac event (MACE), and moderate stenosis was a risk factor of both 5-year MACE and 5-year recurrent angina. In women with CAS, mild stenosis was a risk factor of 5-year MACE, while myocardial bridge was a risk factor of 5-year recurrent angina. Conclusions: In this study, there were sex differences in the angiographic and clinical parameters during the acetylcholine provocation test, incidence of CAS, risk factors of CAS, 5-year MACE, and recurrent angina. Old age, dyslipidemia, and MB were independent risk factors of CAS in both sexes. However, major clinical outcomes up to five years in CAS patients were not different according to sex.

Real-World Three-Year Clinical Outcomes of Biolimus-Eluting Stents versus Other Contemporary Drug-Eluting Stents in Patients with Acute Myocardial Infarction Patients: Data from the Korea Acute Myocardial Infarction Registry (KAMIR).

INTRODUCTION: Biolimus-eluting stents (BES) are known to be superior to bare-metal stents. This study aims to evaluate the safety and efficacy of BES compared to other drug-eluting stents (DES) based on big data from the Korea Acute Myocardial Infarction Registry (KAMIR). METHODS: The study analyzed a total of 9,759 acute myocardial infarction (AMI) patients who underwent percutaneous coronary intervention (PCI) with DES. Total death, cardiac death, recurrent MI, revascularization, stent thrombosis, target lesion failure (TLF, composite of cardiac death, recurrent myocardial infarction (MI), and target lesion revascularization), and major adverse cardiac events (MACE, composite of total death, recurrent MI, and revascularization) were analyzed in patients with AMI up to three years. Study populations were divided into BES (n = 2,020), everolimus-eluting stents (EES, n = 5,293), and zotarolimus-eluting stents (ZES, n = 2,446) groups. RESULTS: To adjust baseline potential confounders, an inverse probability weighting (IPTW) analysis was performed. After IPTW, at three years, total death (7.2%, 8.6%, and 9.5%, P < 0.001), cardiac death (4.1%, 5.3%, and 6.6%, P < 0.001), recurrent MI (1.6%, 2.6%, and 3.2%, P < 0.001), TLF (6.5%, 8.1%, and 9.1%, P < 0.001), and MACE (15.8%, 17.5%, and 18.2%, P < 0.001) were lowest in the BES group compared with the other DES groups in AMI patients. During the 3-year clinical follow-up, the BES group showed better outcomes of MACE (hazard ratio (HR), 0.773; 95% confidence interval (CI), 0.676-0.884; P < 0.001), TLF (HR, 0.659; 95% CI, 0.538-0.808; P < 0.001), total death (HR, 0.687; 95% CI, 0.566-0.835; P < 0.001), and cardiac death (HR,0.593; 95% CI, 0.462-0.541; P < 0.001) than the EES groups. CONCLUSIONS: In this study, BES was superior to EES or ZES in reducing total death, cardiac death, TLF, and MACE in AMI patients.

A Machine Learning-Based Approach for the Prediction of Acute Coronary Syndrome Requiring Revascularization.

The aim of this study is to predict acute coronary syndrome (ACS) requiring revascularization in those patients presenting early-stage angina-like symptom using machine learning algorithms. We obtained data from 2344 ACS patients, who required revascularization and from 3538 non-ACS patients. We analyzed 20 features that are relevant to ACS using standard algorithms, support vector machines and linear discriminant analysis. Based on feature pattern and filter characteristics, we analyzed and extracted a strong prediction function out of the 20 selected features. The obtained prediction functions are relevant showing the area under curve of 0.860 for the prediction of ACS that requiring revascularization. Some features are missing in many data though they are considered to be very informative; it turned out that omitting those features from the input and using more data without those features for training improves the prediction accuracy. Additionally, from the investigation using the receiver operating characteristic curves, a reliable prediction of 2.60% of non-ACS patients could be made with a specificity of 1.0. For those 2.60% non-ACS patients, we can consider the recommendation of medical treatment without risking misdiagnosis of the patients requiring revascularization. We investigated prediction algorithm to select ACS patients requiring revascularization and non-ACS patients presenting angina-like symptoms at an early stage. In the future, a large cohort study is necessary to increase the prediction accuracy and confirm the possibility of safely discriminating the non-ACS patients from the ACS patients with confidence.

Bias Reduction and Metric Learning for Nearest-Neighbor Estimation of Kullback-Leibler Divergence.

Nearest-neighbor estimators for the Kullback-Leiber (KL) divergence that are asymptotically unbiased have recently been proposed and demonstrated in a number of applications. However, with a small number of samples, nonparametric methods typically suffer from large estimation bias due to the nonlocality of information derived from nearest-neighbor statistics. In this letter, we show that this estimation bias can be mitigated by modifying the metric function, and we propose a novel method for learning a locally optimal Mahalanobis distance function from parametric generative models of the underlying density distributions. Using both simulations and experiments on a variety of data sets, we demonstrate that this interplay between approximate generative models and nonparametric techniques can significantly improve the accuracy of nearest-neighbor-based estimation of the KL divergence.

Hyperuricaemia and development of type 2 diabetes mellitus in Asian population.

Recently, meta-analysis studies reported that hyperuricaemia is associated with higher incidence of type 2 diabetes mellitus (T2DM), however, there are limited data on the Asian population. The aim of this observational study is to estimate the long-term impact of hyperuricaemia on the new-onset T2DM and cardiovascular events. This study is based on a single-centre, all-comers, and large retrospective cohort. Subjects that visited from January 2004 to February 2014 were enrolled using the electronic database of Korea University Guro Hospital. A total of 10 505 patients without a history of T2DM were analyzed for uric acid, fasting glucose and haemoglobin (Hb) A1c level. Inclusion criteria included both Hb A1c <5.7% and fasting glucose level <100 mg/dL without T2DM. Hyperuricaemia was defined as a uric acid level ≥7.0 mg/dL in men, and ≥6.5 mg/dL in women. To adjust baseline confounders, a propensity score matching (PSM) analysis was performed. The impact of hyperuricaemia on the new-onset T2DM and cardiovascular events were compared with the non-hyperuricaemia during the 5-year clinical follow-up. After PSM, baseline characteristics of both groups were balanced. In a 5-year follow-up, the hyperuricaemia itself was a strong independent predictor of the incidence of new-onset T2DM (HR, 1.78; 95% CI, 1.12 to 2.8). Hyperuricaemia was a strong independent predictor of new-onset T2DM, which suggests a substantial implication for a correlation between uric acid concentration and insulin resistance (or insulin sensitivity). Also, hyperuricaemia is substantially implicated in cardiovascular risks and the further long-term cardiovascular events in the crude population, but it is not an independent predictor of long-term cardiovascular mortality in the matched population.

Direct Density Derivative Estimation.

Estimating the derivatives of probability density functions is an essential step in statistical data analysis. A naive approach to estimate the derivatives is to first perform density estimation and then compute its derivatives. However, this approach can be unreliable because a good density estimator does not necessarily mean a good density derivative estimator. To cope with this problem, in this letter, we propose a novel method that directly estimates density derivatives without going through density estimation. The proposed method provides computationally efficient estimation for the derivatives of any order on multidimensional data with a hyperparameter tuning method and achieves the optimal parametric convergence rate. We further discuss an extension of the proposed method by applying regularized multitask learning and a general framework for density derivative estimation based on Bregman divergences. Applications of the proposed method to nonparametric Kullback-Leibler divergence approximation and bandwidth matrix selection in kernel density estimation are also explored.

Association of inflammation, myocardial fibrosis and cardiac remodelling in patients with mild aortic stenosis as assessed by biomarkers and echocardiography.

1. The aim of the present study was to investigate the relationships among inflammation, myocardial fibrosis and cardiac remodelling in patients with mild aortic stenosis (AS), as assessed by biomarkers and echocardiography. 2. We evaluated 32 consecutive patients with mild AS, as well as 30 age- and gender-matched healthy individuals with normal aortic valves as control subjects. 3. Baseline echocardiography showed that the left ventricular (LV) mass index (111.3 ± 26.9 vs 94.5 ± 18.2 g/m(2); P = 0.006) and left atrial (LA) volume index (LAVI 27.5 ± 9.0 vs xx.x ± 5.2 mm(3)/mm(2); P = 0.005) were significantly higher in patients with mild AS. 4. Furthermore, LA enlargement (LAVI > 33 mm(3)/mm(2); 32.4% vs 3.3%; P = 0.003) and elevated LV filling pressure (E/e' > 15; 50.0% vs 23.3%; P = 0.036) were higher in patients with mild AS. 5. In patients with mild AS, stepwise, multivariate linear regression analysis revealed that the LV end-diastolic volume index was independently associated with matrix metalloproteinase (MMP)-1 (ß = 0.371; P = 0.015), that the aortic valve mean pressure gradient was independently associated with MMP-2 (ß = 0.19; P = 0.019), that MMP-2 was independently associated with transforming growth factor-ß (ß = 0.95; P < 0.001) and interleukin (IL)-1 (ß = 0.17; P = 0.019) and that IL-1 was independently associated with tissue inhibitor of matrix metalloproteinase-1 (ß = 0.68; P = 0.001). 6. Myocardial fibrosis in mild AS is independently associated with three factors: LV volume overload, aortic valve pressure overload and inflammation.

Prediction of xerostomia in elderly based on clinical characteristics and salivary flow rate with machine learning.

Xerostomia may be accompanied by changes in salivary flow rate and the incidence increases in elderly. We aimed to use machine learning algorithms, to identify significant predictors for the presence of xerostomia. This study is the first to predict xerostomia with salivary flow rate in elderly based on artificial intelligence. In a cross-sectional study, 829 patients with oral discomfort were enrolled, and six features (sex, age, unstimulated and stimulated salivary flow rates (UFR and SFR, respectively), number of systemic diseases, and medication usage) were used in four machine learning algorithms to predict the presence of xerostomia. The incidence of xerostomia increased with age. The SFR was significantly higher than the UFR, and the UFR and SFR were significantly correlated. The UFR, but not SFR, decreased with age significantly. In patients more than 60 years of age, the UFR had a significantly higher predictive accuracy for xerostomia than the SFR. Using machine learning algorithms with tenfold cross-validation, the prediction accuracy increased significantly. In particular, the prediction accuracy of the multilayer perceptron (MLP) algorithm that combined UFR and SFR data was significantly better than either UFR or SFR individually. Moreover, when sex, age, number of systemic diseases, and number of medications were added to the MLP model, the prediction accuracy increased from 56 to 68%.

CLDN18: Clinical, Pathological, and Genetic Signatures with Drug Screening in Gastric Adenocarcinoma.

INTRODUCTION: The CLDN18 gene, encoding claudin 18.1 and claudin 18.2, is a key component of tight junction strands in epithelial cells that form a paracellular barrier that is critical in Stomach Adenocarcinoma (STAD). METHODS: Our study included 1,095 patients with proven STAD, 415 from The Cancer Genome Atlas (TCGA) cohort and 680 from the Gene Expression Omnibus database. We applied various analyses, including gene set enrichment analysis, pathway analysis, and in vitro drug screening to evaluate survival, immune cells, and genes and gene sets associated with cancer progression, based on CLDN18 expression levels. Gradient boosting machine learning (70% for training, 15% for validation, and 15% for testing) was used to evaluate the impact of CLDN18 on survival and develop a survival prediction model. RESULTS: High CLDN18 expression correlated with worse survival in lymphocyte-poor STAD, accompanied by decreased helper T cells, altered metabolic genes, low necrosis-related gene expression, and increased tumor proliferation. CLDN18 expression showed associations with gene sets associated with various stomach, breast, ovarian, and esophageal cancers, while pathway analysis linked CLDN18 to immunity. Incorporating CLDN18 expression improved survival prediction in a machine learning model. Notably, nutlin-3a and niraparib effectively inhibited high CLDN18-expressing gastric cancer cells in drug screening. CONCLUSION: Our study provides a comprehensive understanding of the biological role of CLDN18-based bioinformatics and machine learning analysis in STAD, shedding light on its prognostic significance and potential therapeutic implications. To fully elucidate the molecular intricacies of CLDN18, further investigation is warranted, particularly through in vitro and in vivo studies.

Glycogen synthase kinase 3 activity enhances liver inflammation in MASH.

Background & Aims: Metabolic dysfunction-associated steatohepatitis (MASH) is characterized by excessive circulating toxic lipids, hepatic steatosis, and liver inflammation. Monocyte adhesion to liver sinusoidal endothelial cells (LSECs) and transendothelial migration (TEM) are crucial in the inflammatory process. Under lipotoxic stress, LSECs develop a proinflammatory phenotype known as endotheliopathy. However, mediators of endotheliopathy remain unclear. Methods: Primary mouse LSECs isolated from C57BL/6J mice fed chow or MASH-inducing diets rich in fat, fructose, and cholesterol (FFC) were subjected to multi-omics profiling. Mice with established MASH resulting from a choline-deficient high-fat diet (CDHFD) or FFC diet were also treated with two structurally distinct GSK3 inhibitors (LY2090314 and elraglusib [9-ING-41]). Results: Integrated pathway analysis of the mouse LSEC proteome and transcriptome indicated that leukocyte TEM and focal adhesion were the major pathways altered in MASH. Kinome profiling of the LSEC phosphoproteome identified glycogen synthase kinase (GSK)-3ß as the major kinase hub in MASH. GSK3ß-activating phosphorylation was increased in primary human LSECs treated with the toxic lipid palmitate and in human MASH. Palmitate upregulated the expression of C-X-C motif chemokine ligand 2, intracellular adhesion molecule 1, and phosphorylated focal adhesion kinase, via a GSK3-dependent mechanism. Congruently, the adhesive and transendothelial migratory capacities of primary human neutrophils and THP-1 monocytes through the LSEC monolayer under lipotoxic stress were reduced by GSK3 inhibition. Treatment with the GSK3 inhibitors LY2090314 and elraglusib ameliorated liver inflammation, injury, and fibrosis in FFC- and CDHFD-fed mice, respectively. Immunophenotyping using cytometry by mass cytometry by time of flight of intrahepatic leukocytes from CDHFD-fed mice treated with elraglusib showed reduced infiltration of proinflammatory monocyte-derived macrophages and monocyte-derived dendritic cells. Conclusion: GSK3 inhibition attenuates lipotoxicity-induced LSEC endotheliopathy and could serve as a potential therapeutic strategy for treating human MASH. Impact and Implications: LSECs under lipotoxic stress in MASH develop a proinflammatory phenotype known as endotheliopathy, with obscure mediators and functional outcomes. The current study identified GSK3 as the major driver of LSEC endotheliopathy, examined its pathogenic role in myeloid cell-associated liver inflammation, and defined the therapeutic efficacy of pharmacological GSK3 inhibitors in murine MASH. This study provides preclinical data for the future investigation of GSK3 pharmacological inhibitors in human MASH. The results of this study are important to hepatologists, vascular biologists, and investigators studying the mechanisms of inflammatory liver disease and MASH, as well as those interested in drug development.

Glycolysis in hepatic stellate cells coordinates fibrogenic extracellular vesicle release spatially to amplify liver fibrosis.

Liver fibrosis is characterized by the activation of perivascular hepatic stellate cells (HSCs), the release of fibrogenic nanosized extracellular vesicles (EVs), and increased HSC glycolysis. Nevertheless, how glycolysis in HSCs coordinates fibrosis amplification through tissue zone-specific pathways remains elusive. Here, we demonstrate that HSC-specific genetic inhibition of glycolysis reduced liver fibrosis. Moreover, spatial transcriptomics revealed a fibrosis-mediated up-regulation of EV-related pathways in the liver pericentral zone, which was abrogated by glycolysis genetic inhibition. Mechanistically, glycolysis in HSCs up-regulated the expression of EV-related genes such as Ras-related protein Rab-31 (RAB31) by enhancing histone 3 lysine 9 acetylation on the promoter region, which increased EV release. Functionally, these glycolysis-dependent EVs increased fibrotic gene expression in recipient HSC. Furthermore, EVs derived from glycolysis-deficient mice abrogated liver fibrosis amplification in contrast to glycolysis-competent mouse EVs. In summary, glycolysis in HSCs amplifies liver fibrosis by promoting fibrogenic EV release in the hepatic pericentral zone, which represents a potential therapeutic target.

Pre-test probability for coronary artery disease in patients with chest pain based on machine learning techniques.

BACKGROUND: A correct and prompt diagnosis of coronary artery disease (CAD) is a crucial component of disease management to reduce the risk of death and improve the quality of life in patients with CAD. Currently, the American College of Cardiology (ACC)/American Heart Association (AHA) and the European Society of Cardiology (ESC) guidelines recommend selecting an appropriate pre-diagnosis test for an individual patient according to the CAD probability. The purpose of this study was to develop a practical pre-test probability (PTP) for obstructive CAD in patients with chest pain using machine learning (ML); also, the performance of ML-PTP for CAD is compared to the final result of coronary angiography (CAG). METHODS: We used a database from a single-center, prospective, all-comer registry designed to reflect real-world practice since 2004. All subjects underwent invasive CAG at Korea University Guro Hospital in Seoul, South Korea. We used logistic regression algorithms, random forest (RF), supporting vector machine, and K-nearest neighbor classification for the ML models. The dataset was divided into two consecutive sets according to the registration period to validate the ML models. ML training for PTP and internal validation used the first dataset registered between 2004 and 2012 (8631 patients). The second dataset registered between 2013 and 2014 (1546 patients) was used for external validation. The primary endpoint was obstructive CAD. Obstructive CAD was defined as having a stenosis diameter of >70% on the quantitative CAG of the main epicardial coronary artery. RESULTS: We derived an ML-based model consisting of three different models according to the subject used to obtain the information, such as the patient himself (dataset 1), the community's first medical center (dataset 2), and doctors (dataset 3). The performance range of the ML-PTP models as the non-invasive test had C-statistics of 0.795 to 0.984 compared to the result of invasive testing via CAG in patients with chest pain. The training ML-PTP models were adjusted to have 99% sensitivity for CAD so as not to miss actual CAD patients. In the testing dataset, the best accuracy of the ML-PTP model was 45.7% using dataset 1, 47.2% using dataset 2, and 92.8% using dataset 3 and the RF algorithm. The CAD prediction sensitivity was 99.0%, 99.0%, and 98.0%, respectively. CONCLUSION: We successfully developed a high-performance model of ML-PTP for CAD which is expected to reduce the need for non-invasive tests in chest pain. However, since this PTP model is derived from data of a single medical center, multicenter verification is required to use it as a PTP recommended by the major American societies and the ESC.

Low CDKN1B Expression Associated with Reduced CD8+ T Lymphocytes Predicts Poor Outcome in Breast Cancer in a Machine Learning Analysis.

The cyclin-dependent kinase inhibitor 1B (CDKN1B) gene, which encodes the p27Kip1 protein, is important in regulating the cell cycle process and cell proliferation. Its role in breast cancer prognosis is controversial. We evaluated the significance and predictive role of CDKN1B expression in breast cancer prognosis. We investigated the clinicopathologic factors, survival rates, immune cells, gene sets, and prognostic models according to CDKN1B expression in 3794 breast cancer patients. We performed gene set enrichment analysis (GSEA), in silico cytometry, pathway network analyses, gradient boosting machine (GBM) learning, and in vitro drug screening. High CDKN1B expression levels in breast cancer correlated with high lymphocyte infiltration signature scores and increased CD8+ T cells, both of which were associated with improved prognosis in breast cancer. which were associated with a better prognosis. CDKN1B expression was associated with gene sets for the upregulation of T-cell receptor signaling pathways and downregulation of CD8+ T cells. Pathway network analysis revealed a direct link between CDKN1B and the pathway involved in the positive regulation of the protein catabolic process pathway. In addition, an indirect link was identified between CDKN1B and the T-cell receptor signaling pathway. In in vitro drug screening, BMS-345541 demonstrated efficacy as a therapeutic targeting of CDKN1B, effectively impeding the growth of breast cancer cells characterized by low CDKN1B expression. The inclusion of CDKN1B expression in GBM models increased the accuracy of survival predictions. CDKN1B expression plays a significant role in breast cancer progression, implying that targeting CDKN1B might be a promising strategy for treating breast cancer.

Low gamma-butyrobetaine dioxygenase (BBOX1) expression as a prognostic biomarker in patients with clear cell renal cell carcinoma: a machine learning approach.

Gamma-butyrobetaine dioxygenase (BBOX1) is a catalyst for the conversion of gamma-butyrobetaine to l-carnitine, which is detected in normal renal tubules. The purpose of this study was to analyze the prognosis, immune response, and genetic alterations associated with low BBOX1 expression in patients with clear cell renal cell carcinoma (RCC). We analyzed the relative influence of BBOX1 on survival using machine learning and investigated drugs that can inhibit renal cancer cells with low BBOX1 expression. We analyzed clinicopathologic factors, survival rates, immune profiles, and gene sets according to BBOX1 expression in a total of 857 patients with kidney cancer from the Hanyang University Hospital cohort (247 cases) and The Cancer Genome Atlas (610 cases). We employed immunohistochemical staining, gene set enrichment analysis, in silico cytometry, pathway network analyses, in vitro drug screening, and gradient boosting machines. BBOX1 expression in RCC was decreased compared with that in normal tissues. Low BBOX1 expression was associated with poor prognosis, decreased CD8+ T cells, and increased neutrophils. In gene set enrichment analyses, low BBOX1 expression was related to gene sets with oncogenic activity and a weak immune response. In pathway network analysis, BBOX1 was linked to regulation of various T cells and programmed death-ligand 1. In vitro drug screening showed that midostaurin, BAY-61-3606, GSK690693, and linifanib inhibited the growth of RCC cells with low BBOX1 expression. Low BBOX1 expression in patients with RCC is related to short survival time and reduced CD8+ T cells; midostaurin, among other drugs, may have enhanced therapeutic effects in this context.

Advantages of deep learning with convolutional neural network in detecting disc displacement of the temporomandibular joint in magnetic resonance imaging.

This study investigated the usefulness of deep learning-based automatic detection of anterior disc displacement (ADD) from magnetic resonance imaging (MRI) of patients with temporomandibular joint disorder (TMD). Sagittal MRI images of 2520 TMJs were collected from 861 men and 399 women (average age 37.33 ± 18.83 years). A deep learning algorithm with a convolutional neural network was developed. Data augmentation and the Adam optimizer were applied to reduce the risk of overfitting the deep-learning model. The prediction performances were compared between the models and human experts based on areas under the curve (AUCs). The fine-tuning model showed excellent prediction performance (AUC = 0.8775) and acceptable accuracy (approximately 77%). Comparing the AUC values of the from-scratch (0.8269) and freeze models (0.5858) showed lower performances of the other models compared to the fine-tuning model. In Grad-CAM visualizations, the fine-tuning scheme focused more on the TMJ disc when judging ADD, and the sparsity was higher than that of the from-scratch scheme (84.69% vs. 55.61%, p < 0.05). The three fine-tuned ensemble models using different data augmentation techniques showed a prediction accuracy of 83%. Moreover, the AUC values of ADD were higher when patients with TMD were divided by age (0.8549-0.9275) and sex (male: 0.8483, female: 0.9276). While the accuracy of the ensemble model was higher than that of human experts, the difference was not significant (p = 0.1987-0.0671). Learning from pre-trained weights allowed the fine-tuning model to outperform the from-scratch model. Another benefit of the fine-tuning model for diagnosing ADD of TMJ in Grad-CAM analysis was the deactivation of unwanted gradient values to provide clearer visualizations compared to the from-scratch model. The Grad-CAM visualizations also agreed with the model learned through important features in the joint disc area. The accuracy was further improved by an ensemble of three fine-tuning models using diversified data. The main benefits of this model were the higher specificity compared to human experts, which may be useful for preventing true negative cases, and the maintenance of its prediction accuracy across sexes and ages, suggesting a generalized prediction.

Age group prediction with panoramic radiomorphometric parameters using machine learning algorithms.

The aim of this study is to investigate the relationship of 18 radiomorphometric parameters of panoramic radiographs based on age, and to estimate the age group of people with permanent dentition in a non-invasive, comprehensive, and accurate manner using five machine learning algorithms. For the study population (209 men and 262 women; mean age, 32.12 ± 18.71 years), 471 digital panoramic radiographs of Korean individuals were applied. The participants were divided into three groups (with a 20-year age gap) and six groups (with a 10-year age gap), and each age group was estimated using the following five machine learning models: a linear discriminant analysis, logistic regression, kernelized support vector machines, multilayer perceptron, and extreme gradient boosting. Finally, a Fisher discriminant analysis was used to visualize the data configuration. In the prediction of the three age-group classification, the areas under the curve (AUCs) obtained for classifying young ages (10-19 years) ranged from 0.85 to 0.88 for five different machine learning models. The AUC values of the older age group (50-69 years) ranged from 0.82 to 0.88, and those of adults (20-49 years) were approximately 0.73. In the six age-group classification, the best scores were also found in age groups 1 (10-19 years) and 6 (60-69 years), with mean AUCs ranging from 0.85 to 0.87 and 80 to 0.90, respectively. A feature analysis based on LDA weights showed that the L-Pulp Area was important for discriminating young ages (10-49 years), and L-Crown, U-Crown, L-Implant, U-Implant, and Periodontitis were used as predictors for discriminating older ages (50-69 years). We established acceptable linear and nonlinear machine learning models for a dental age group estimation using multiple maxillary and mandibular radiomorphometric parameters. Since certain radiomorphological characteristics of young and the elderly were linearly related to age, young and old groups could be easily distinguished from other age groups with automated machine learning models.