A5103019877- Coronary Interventions and Diagnostics
- Cardiac Imaging and Diagnostics
- Acute Myocardial Infarction Research
- Machine Learning in Bioinformatics
- Protein Structure and Dynamics
- Ferroptosis and cancer prognosis
- Cerebrovascular and Carotid Artery Diseases
- Bioinformatics and Genomic Networks
- Optical Coherence Tomography Applications
- Lung Cancer Treatments and Mutations
- Enzyme Structure and Function
- Antiplatelet Therapy and Cardiovascular Diseases
- RNA and protein synthesis mechanisms
- Epigenetics and DNA Methylation
- Genomics and Phylogenetic Studies
- COVID-19 Clinical Research Studies
- Anomaly Detection Techniques and Applications
- Network Security and Intrusion Detection
- Pituitary Gland Disorders and Treatments
- Coronary Artery Anomalies
- Fractal and DNA sequence analysis
- Integrated Circuits and Semiconductor Failure Analysis
- Advanced Computing and Algorithms
- Protease and Inhibitor Mechanisms
- Advanced MRI Techniques and Applications
Harbin Medical University
2017-2025
Second Affiliated Hospital of Harbin Medical University
2020-2025
Ministry of Education of the People's Republic of China
2019-2025
Beijing University of Civil Engineering and Architecture
2024
AstraZeneca (Brazil)
2024
AstraZeneca (United States)
2019-2024
Affiliated Hospital of Hebei University
2023-2024
Creative Commons
2022
Multichannel (China)
2022
Heilongjiang University of Chinese Medicine
2022
Abstract Protein secondary structure prediction can provide important information for protein 3D and functions. Deep learning offers a new opportunity to significantly improve accuracy. In this article, deep neural network architecture, named the inception‐inside‐inception (Deep3I) network, is proposed implemented as software tool MUFOLD‐SS. The input MUFOLD‐SS carefully designed feature matrix corresponding primary amino acid sequence of protein, which consists rich set derived from...
Abstract Background Several computational methods have been developed to predict protein-protein interactions from amino acid sequences, but most of those are intended for the within a species rather than across different species. Methods predicting between homogeneous proteins not appropriate finding heterogeneous since they do distinguish same Results We new method representing protein sequence variable length in frequency vector fixed length, which encodes relative three consecutive acids...
Osteoarthritis (OA) is the most common type of arthritis with no effective therapy. Subchondral bone and overlying articular cartilage are closely associated function as "osteo-chondral unit" in joint. Abnormal mechanical load leads to activated osteoclast activity increased resorption subchondral bone, which implicated onset OA pathogenesis. Thus, inhibiting activation could prevent onset. Betaine, isolated from Lycii Radicis Cortex (LRC), has been demonstrated exert anti-inflammatory,...
Hybrid intravascular ultrasound-optical coherence tomography (IVUS-OCT) imaging can integrate both advantages. Optical ultrasonic flow ratio (OUFR) was recently developed for functional assessment. This study aimed to verify the diagnostic performance of OUFR using fractional reserve (FFR) as reference standard. From February 2023 June 2023, 80 patients (126 vessels) with unstable angina pectoris were assessed FFR and either hybrid IVUS-OCT or stand-alone IVUS before percutaneous coronary...
Previous studies have reported the value of quantitative flow ratio (QFR) to assess physiological significance non-culprit lesions (NCLs) in acute myocardial infarction (AMI) patients and optical coherence tomography (OCT)-defined thin-cap fibroatheroma (TCFA) identify vulnerable plaques. We sought systematically compare long-term NCL-related clinical prognosis an AMI population utilising Murray fractal law-based QFR (μQFR) values OCT-defined TCFA. Three-vessel OCT imaging μQFR assessment...
RNA is emerging as a valuable target for the development of novel therapeutic agents. The rational design RNA-targeting small molecules, however, has been hampered by relative lack methods analysis molecule–RNA interactions. Here, we present our efforts to develop such platform using photoaffinity labeling. This technique, termed Photoaffinity Evaluation Ligation-Sequencing (PEARL-seq), enables rapid identification molecule binding locations within their targets and can provide information...
Abstract Immuno-oncology (IO) therapies have transformed the therapeutic landscape of non-small cell lung cancer (NSCLC). However, patient responses to IO are variable and influenced by a heterogeneous combination health, immune, tumor factors. There is pressing need discover distinct NSCLC subgroups that influence response. We developed dee p pa tient graph convolutional n etwork, we call “DeePaN”, complexity across data modalities impacting benefit. DeePaN employs high-dimensional derived...
Prediction of protein backbone torsion angles (Psi and Phi) can provide important information for structure prediction sequence alignment. Existing methods Psi-Phi angle have significant room improvement. In this paper, a new deep residual inception network architecture, called DeepRIN, is proposed the angles. The input to DeepRIN feature matrix representing composition physico-chemical properties amino acids, 20-dimensional position-specific substitution (PSSM) generated by PSI-BLAST,...
The cell-mediated protective and pathogenic immune responses to SARS-CoV-2 infection remain largely elusive. Here we identified 76 distinct cell subsets in the PBMC samples that were associated with various clinical presentations of COVID-19 using scRNA-seq technology coupled a deep comprehensive analysis unique surface markers differentially expressed genes. We revealed (TRAV1-2 + CD8 )MAIT cells (NCAM1 hi CD160 )NK significantly enriched asymptomatic subjects whereas (LAG3 )NKT increased...
ObjectivesThis study sought to explore the use of novel natural language processing (NLP) methods for classifying unstructured, qualitative textual data from interviews patients with cancer identify patient-reported symptoms and impacts on quality life.MethodsWe tested ability 4 NLP models accurately classify text interview transcripts as "symptom," "quality life impact," "other." Interview sets hepatocellular carcinoma (HCC) (n = 25), biliary tract (BTC) 23), gastric 24) were used. Models...
Patients with acute myocardial infarction and angiographically obstructive non-culprit lesions are at high risk for recurrent major adverse cardiac events (MACEs). However, it remains largely unknown whether due to stenosis severity or the underlying high-risk lesion morphology. Between January 2017 December 2021, 1312 patients underwent optical coherence tomography of all 3 main epicardial arteries after successful percutaneous coronary intervention. were categorized according presence...
Abstract Immunotherapy (IO) has revolutionized the management of non-small-cell lung cancer (NSCLC), leading to groundbreaking improvements in patient survival. However, resistance IO some patients remains a challenge. Existing biomarkers are based on either clinical or genomic factors. Combining factors with health characteristics is pivotal developing new markers and improving accuracy predictions treatment resistance. The objective this study develop data-driven model no priori...
Abstract Protein gamma-turn prediction is useful in protein function studies and experimental design. Several methods for have been developed, but the results were unsatisfactory with Matthew correlation coefficients (MCC) around 0.2–0.4. Hence, it worthwhile exploring new prediction. A cutting-edge deep neural network, named Capsule Network (CapsuleNet), provides a opportunity Even when number of input samples relatively small, capsules from CapsuleNet are effective to extract high-level...
Subclinical atherothrombosis and plaque healing may lead to rapid progression. The histopathologic healed has a layered appearance when imaged using optical coherence tomography. We assessed the frequency, predictors, distribution, morphological characteristics of tomography culprit nonculprit plaques in patients with acute myocardial infarction.A prospective series 325 infarction underwent imaging all 3 native coronary arteries. Layered phenotype had heterogeneous signal-rich tissue located...
Abstract The precise understanding of hypothalamic injury (HI) patterns and their relationship with different craniopharyngioma (CP) classifications remains poorly addressed. Here, four HI after CP resection based on endoscopic observation were introduced. A total 131 cases treated endonasal approach (EEA) reviewed retrospectively divided into patterns: no‐HI, mild‐HI, unilateral‐HI bilateral‐HI, according to intraoperative findings. outcomes evaluated compared between groups in terms weight...
Background: Previous studies have found that coronary artery calcification is closely associated with the occurrence of major adverse cardiac events (MACE). This study aimed to investigate characteristics and clinical outcomes different calcified plaques in patients acute syndrome (ACS) by using optical coherence tomography (OCT). Methods: 258 ACS culprit who underwent OCT-guided stent implantation were enrolled. They divided into three subtypes based on plaque morphology, including eruptive...
To analyze plaque characteristics of non-culprit coronary lesions with cholesterol crystals in patients acute myocardial infarction(AMI) by using optical coherence tomography(OCT). We also investigated the potential association between rupture and healed at segment.
Abstract Beta‐turn prediction is useful in protein function studies and experimental design. Although recent approaches using machine‐learning techniques such as support vector machine (SVM), neural networks, K nearest neighbor have achieved good results for beta‐turn prediction, there still significant room improvement. As previous predictors utilized features a sliding window of 4‐20 residues to capture interactions among sequentially neighboring residues, feature engineering may result...