A5066888888- Bioinformatics and Genomic Networks
- Bayesian Modeling and Causal Inference
- Gene expression and cancer classification
- AI in cancer detection
- Biomedical Text Mining and Ontologies
- Genetic Associations and Epidemiology
- Data-Driven Disease Surveillance
- Epigenetics and DNA Methylation
- Radiomics and Machine Learning in Medical Imaging
- Diabetes Treatment and Management
- Anomaly Detection Techniques and Applications
- Cancer-related gene regulation
- Histone Deacetylase Inhibitors Research
- Metabolism, Diabetes, and Cancer
- FOXO transcription factor regulation
- RNA modifications and cancer
- Bayesian Methods and Mixture Models
- MicroRNA in disease regulation
- Diabetes Management and Research
- Data Mining Algorithms and Applications
- Ubiquitin and proteasome pathways
- Ferroptosis and cancer prognosis
- Artificial Intelligence in Healthcare
- Diet and metabolism studies
- Pancreatic function and diabetes
University of Pittsburgh
2015-2025
Tianjin Medical University
2011-2024
Tianjin First Center Hospital
2011-2024
Sichuan Mianyang 404 Hospital
2024
Zhengzhou University of Aeronautics
2023
Zhengzhou University
2023
Nankai University
2023
Genome Institute of Singapore
2006-2019
Center for Discovery
2018-2019
Agency for Science, Technology and Research
2008-2019
Polycomb-repressive complex 2 (PRC2)-mediated histone methylation plays an important role in aberrant cancer gene silencing and is a potential target for therapy. Here we show that S-adenosylhomocysteine hydrolase inhibitor 3-Deazaneplanocin A (DZNep) induces efficient apoptotic cell death cells but not normal cells. We found DZNep effectively depleted cellular levels of PRC2 components EZH2, SUZ12, EED inhibited associated H3 Lys 27 (but 9 methylation). By integrating RNA interference...
Inhibitors of histone deacetylases (HDACIs) are a new generation anticancer agents that selectively kill tumor cells. However, the molecular basis for their selectivity is not well understood. We investigated effects HDACIs on oncogenic Rb-E2F1 pathway, which frequently deregulated in human cancers. Here, we report cancer cells with elevated E2F1 activity, caused either by enforced expression, or E1A oncogene highly susceptible to HDACI-induced cell death. This E2F1-mediated apoptosis...
The Rb–E2F pathway drives cell cycle progression and proliferation, the molecular strategies safeguarding its activity are not fully understood. Here we report that E2F1 directly transactivates miR-449a/b . targets inhibits oncogenic CDK6 CDC25A , resulting in pRb dephosphorylation arrest at G1 phase, revealing a negative feedback regulation of pRb–E2F1 pathway. Moreover, miR-449 a/b expression cancer cells is epigenetically repressed through histone H3 Lys27 trimethylation, epigenetic drug...
Epithelial-mesenchymal transition (EMT) in cancer cells plays a pivotal role determining metastatic prowess, but knowledge of EMT regulation remains incomplete. In this study, we defined critical functional for the Forkhead transcription factor FOXQ1 regulating breast cells. expression was correlated with high-grade basal-like cancers and associated poor clinical outcomes. RNAi-mediated suppression highly invasive human reversed EMT, reduced ability, alleviated other aggressive phenotypes...
It is important to be able predict, for each individual patient, the likelihood of later metastatic occurrence, because prediction can guide treatment plans tailored a specific patient prevent metastasis and help avoid under-treatment or over-treatment. Deep neural network (DNN) learning, commonly referred as deep has become popular due its success in image detection prediction, but questions such whether learning outperforms other machine methods when using non-image clinical data remain...
Gene-gene epistatic interactions likely play an important role in the genetic basis of many common diseases. Recently, machine-learning and data mining methods have been developed for learning relationships from data. A well-known combinatorial method that has successfully applied detecting epistasis is Multifactor Dimensionality Reduction (MDR). Jiang et al. created a called BNMBL to learn Bayesian network (BN) models. They compared MDR using simulated sets. Each these sets was generated...
Identifying local recurrences in breast cancer from patient data sets is important for clinical research and practice. Developing a model using natural language processing machine learning to identify patients can reduce the time-consuming work of manual chart review.We design novel concept-based filter prediction detect EHRs. In training dataset, we manually review development corpus 50 progress notes extract partial sentences that indicate recurrence. We process these obtain set Unified...
Abstract The recent announcement of the Precision Medicine Initiative by President Obama has brought precision medicine (PM) to forefront for healthcare providers, researchers, regulators, innovators, and funders alike. As technologies continue evolve datasets grow in magnitude, a strong computational infrastructure will be essential realize PM’s vision improved derived from personal data. In addition, informatics research innovation affords tremendous opportunity drive science underlying...
Overfitting may affect the accuracy of predicting future data because weakened generalization. In this research, we used an electronic health records (EHR) dataset concerning breast cancer metastasis to study overfitting deep feedforward neural networks (FNNs) prediction models. We studied how each hyperparameter and some interesting pairs hyperparameters were interacting influence model performance overfitting. The 11 activate function, weight initializer, number hidden layers, learning...
Efforts to improve the clinical outcome of highly aggressive triple-negative breast cancer (TNBC) have been hindered by lack effective targeted therapies. Thus, it is important identify specific gene targets/pathways driving invasive phenotype develop more therapeutics. Here we show that ubiquitin-associated and SH3 domain-containing B (UBASH3B), a protein tyrosine phosphatase, overexpressed in TNBC, where supports malignant growth, invasion, metastasis largely through modulating epidermal...
Abstract The transcription factor FOXM1 binds to its consensus sequence at promoters through DNA binding domain (DBD) and activates proliferation-associated genes. aberrant overexpression of correlates with tumorigenesis progression many cancers. Inhibiting transcriptional activities is proposed as a potential therapeutic strategy for cancer treatment. In this study, we obtained FOXM1-specific single stranded aptamer (FOXM1 Apt) by SELEX recombinant DBD protein the target selection. Apt...
Objective A Clinical Decision Support System (CDSS) that can amass Electronic Health Record (EHR) and other patient data holds promise to provide accurate classification guide treatment choices. Our objective is develop the for Making Personalized Assessments Recommendations Concerning Breast Cancer Patients (DPAC), which a CDSS learned from recommends optimal decisions based on patient's features. Method We developed Bayesian network architecture called Causal Modeling with Internal Layers...
Abstract It is believed that interactions among genes (epistasis) may play an important role in susceptibility to common diseases (Moore and Williams [2002]. Ann Med 34:88–95; Ritchie et al. [2001]. Am J Hum Genet 69:138–147). To study the underlying genetic variants of diseases, genome‐wide association studies (GWAS) simultaneously assay several hundreds thousands SNPs are being increasingly used. Often, data from these analyzed with single‐locus methods (Lambert [2009]. Nat 41:1094–1099;...
Abstract Background The problem of learning causal influences from data has recently attracted much attention. Standard statistical methods can have difficulty discrete causes, which interacting to affect a target, because the assumptions in these often do not model relationships well. An important task then is learn such interactions data. Motivated by epistatic datasets developed genome-wide association studies ( GWAS ), researchers conceived new for interactions. However, many...
Cancer is mainly caused by somatic genome alterations (SGAs). Precision oncology involves identifying and targeting tumor-specific aberrations resulting from causative SGAs. We developed a novel computational framework that finds the likely SGAs in an individual tumor estimates their impact on oncogenic processes, which suggests disease mechanisms are acting tumor. This information can be used to guide precision oncology. report causal inference (TCI) framework, modeling relationships...
Background: A grid search, at the cost of training and testing a large number models, is an effective way to optimize prediction performance deep learning models. challenging task concerning search time management. Without good management scheme, can easily be set off as "mission" that will not finish in our lifetime. In this study, we introduce heuristic three-stage mechanism for managing running low-budget searches with learning, sweet-spot (SSGS) randomized (RGS) strategies improving...
A system that monitors a region for disease outbreak is called surveillance system. spatial searches patterns of in subregions the monitored region. temporal looks emerging by analyzing how have changed during recent periods time. If non-spatial, non-temporal could be converted to spatio-temporal one, performance might improved terms early detection, accuracy, and reliability. Bayesian network framework proposed class space-time systems BNST. The applied detection PC order create PCTS....