Search

Abstract This study aimed to evaluate the sensitivity of AI in screening acute leukemia and its capability to classify either physiological or pathological cells. Utilizing an acute leukemia orientation tube (ALOT), one of the protocols of Euroflow, flow cytometry efficiently identifies various forms of acute leukemia. However, the analysis of flow cytometry can be time-consuming work. This retrospective study included 241 patients who underwent flow cytometry examination using ALOT between 2017 and 2022. The collected flow cytometry data were used to train an artificial intelligence using deep learning. The trained AI demonstrated a 94.6% sensitivity in detecting acute myeloid leukemia (AML) patients and a 98.2% sensitivity for B-lymphoblastic leukemia (B-ALL) patients. The sensitivities of physiological cells were at least 80%, with variable performance for pathological cells. In conclusion, the AI, trained with ResNet-50 and EverFlow, shows promising results in identifying patients with AML and B-ALL, as well as classifying physiological cells.

Abstract Polygenic scores estimate genetic susceptibility to diseases. We systematically calculated polygenic scores across 457 phenotypes using genotyping array data from China Medical University Hospital. Logistic regression models assessed polygenic scores’ ability to predict disease traits. The polygenic score model with the highest accuracy, based on maximal area under the receiver operating characteristic curve (AUC), is provided on the GeneAnaBase website of the hospital. Our findings indicate 49 phenotypes with AUC greater than 0.6, predominantly linked to endocrine and metabolic diseases. Notably, hyperplasia of the prostate exhibited the highest disease prediction ability (P value = 1.01 × 10−19, AUC = 0.874), highlighting the potential of these polygenic scores in preventive medicine and diagnosis. This study offers a comprehensive evaluation of polygenic scores performance across diverse human traits, identifying promising applications for precision medicine and personalized healthcare, thereby inspiring further research and development in this field.

Abstract Automated coronary angiography assessment requires precise vessel segmentation, a task complicated by uneven contrast filling and background noise. Our research introduces an ensemble U-Net model, SE-RegUNet, designed to accurately segment coronary vessels using 100 labeled angiographies from angiographic images. SE-RegUNet incorporates RegNet encoders and squeeze-and-excitation blocks to enhance feature extraction. A dual-phase image preprocessing strategy further improves the model's performance, employing unsharp masking and contrast-limited adaptive histogram equalization. Following fivefold cross-validation and Ranger21 optimization, the SE-RegUNet 4GF model emerged as the most effective, evidenced by performance metrics such as a Dice score of 0.72 and an accuracy of 0.97. Its potential for real-world application is highlighted by its ability to process images at 41.6 frames per second. External validation on the DCA1 dataset demonstrated the model's consistent robustness, achieving a Dice score of 0.76 and an accuracy of 0.97. The SE-RegUNet 4GF model's precision in segmenting blood vessels in coronary angiographies showcases its remarkable efficiency and accuracy. However, further development and clinical testing are necessary before it can be routinely implemented in medical practice.

Abstract Large-artery atherosclerosis (LAA) is a leading cause of cerebrovascular disease. However, LAA diagnosis is costly and needs professional identification. Many metabolites have been identified as biomarkers of specific traits. However, there are inconsistent findings regarding suitable biomarkers for the prediction of LAA. In this study, we propose a new method integrates multiple machine learning algorithms and feature selection method to handle multidimensional data. Among the six machine learning models, logistic regression (LR) model exhibited the best prediction performance. The value of area under the receiver operating characteristic curve (AUC) was 0.92 when 62 features were incorporated in the external validation set for the LR model. In this model, LAA could be well predicted by clinical risk factors including body mass index, smoking, and medications for controlling diabetes, hypertension, and hyperlipidemia as well as metabolites involved in aminoacyl-tRNA biosynthesis and lipid metabolism. In addition, we found that 27 features were present among the five adopted models that could provide good results. If these 27 features were used in the LR model, an AUC value of 0.93 could be achieved. Our study has demonstrated the effectiveness of combining machine learning algorithms with recursive feature elimination and cross-validation methods for biomarker identification. Moreover, we have shown that using shared features can yield more reliable correlations than either model, which can be valuable for future identification of LAA.

Abstract Targeting the programmed cell death protein 1/programmed cell death ligand 1 (PD-1/PD-L1) axis with monoclonal antibodies (mAbs) represents a crucial breakthrough in anticancer therapy, but mAbs are limited by their poor oral bioavailability, adverse events in multiple organ systems, and primary, adaptive, and acquired resistance, amongst other issues. More recently, the advent of small molecule inhibitors that target the PD-1/PD-L1 axis have shown promising cellular inhibitory activity and the potential to counteract the disadvantages of mAbs. In this study, structure-based virtual screening identified small molecule inhibitors that effectively inhibited the PD-1/PD-L1 interaction. Six of those small molecule inhibitors were applied to cell-based experiments targeting PD-1: CH-1, CH-2, CH-3, CH-4, CH-5, and CH-6. Of all 6, CH-4 displayed the lowest cytotoxicity and strongest inhibitory activity towards the PD-1/PD-L1 interaction. The experiments revealed that CH-4 inhibited the interaction of soluble form PD-L1 (sPD-L1) with PD-1 surface protein expressed by KG-1 cells. Investigations into CH-4 analogs revealed that CH-4.7 effectively blocked the PD-1/sPD-L1 interaction, but sustained the secretion of interleukin-2 and interferon-γ by Jurkat cells. Our experiments revealed a novel small molecule inhibitor that blocks the interaction of PD-1/sPD-L1 and potentially offers an alternative PD-1 target for immune checkpoint therapy.

Abstract Eradicating influenza A virus (IAV) is difficult, due to its genetic drift and reassortment ability. As the infectious cycle is initiated by the influenza glycoprotein, hemagglutinin (HA), which mediates the binding of virions to terminal sialic acids moieties, HA is a tempting target of anti-influenza inhibitors. However, the complexity of the HA structure has prevented delineation of the structural characterization of the HA protein–ligand complex. Our computational strategy efficiently analyzed > 200,000 records of compounds held in the United States National Cancer Institute (NCI) database and identified potential HA inhibitors, by modeling the sialic acid (SA) receptor binding site (RBS) for the HA structure. Our modeling revealed that compound NSC85561 showed significant antiviral activity against the IAV H1N1 strain with EC50 values ranging from 2.31 to 2.53 µM and negligible cytotoxicity (CC50 > 700 µM). Using the NSC85561 compound as the template to generate 12 derivatives, robust bioassay results revealed the strongest antiviral efficacies with NSC47715 and NSC7223. Virtual screening clearly identified three SA receptor binding site inhibitors that were successfully validated in experimental data. Thus, our computational strategy has identified SA receptor binding site inhibitors against HA that show IAV-associated antiviral activity.

Abstract Histone deacetylases (HDAC) contain eighteen isoforms that can be divided into four classes. Of these isoform enzymes, class IIa (containing HDAC4, 5, 7 and 9) target unique substrates, some of which are client proteins associated with epigenetic control. Class IIa HDACs are reportedly associated with some neuronal disorders, making HDACs therapeutic targets for treating neurodegenerative diseases. Additionally, some reported HDAC inhibitors contain hydroxamate moiety that chelates with zinc ion to become the cofactor of HDAC enzymes. However, the hydroxamate functional group is shown to cause undesirable effects and has poor pharmacokinetic profile. This study used in silico virtual screening methodology to identify several nonhydroxamate compounds, obtained from National Cancer Institute database, which potentially inhibited HDAC4. Comparisons of the enzyme inhibitory activity against a panel of HDAC isoforms revealed these compounds had strong inhibitory activity against class IIa HDACs, but weak inhibitory activity against class I HDACs. Further analysis revealed that a single residue affects the cavity size between class I and class IIa HDACs, thus contributing to the selectivity of HDAC inhibitors discovered in this study. The discovery of these inhibitors presents the possibility of developing new therapeutic treatments that can circumvent the problems seen in traditional hydroxamate-based drugs.