Towards fine-scale population stratification modeling based on kernel principal component analysis and random forest

Zhang Weiwen,Cheng Lianglun,Huang Guoheng 한국유전학회 2021 Genes & Genomics Vol.43 No.10

Background Population stratifcation modeling is essential in Genome-Wide Association Studies. Objective In this paper, we aim to build a fne-scale population stratifcation model to efciently infer individual genetic ancestry. Methods Kernel Principal Component Analysis (PCA) and random forest are adopted to build the population stratifcation model, together with parameter optimization. We explore diferent PCA methods, including standard PCA and kernel PCA to extract relevant features from the genotype data that is transformed by vcf2geno, a pipeline from LASER software. These extracted features are fed into a random forest for ensemble learning. Parameter tuning is performed to jointly fnd the optimal number of principal components, kernel function for PCA and parameters of the random forest. Results Experiments based on HGDP dataset show that kernel PCA with Sigmoid function and Gaussian function can achieve higher prediction accuracy than the standard PCA. Compared to standard PCA with the two principal components, the accuracy by using KPCA-Sigmoid with the optimal number of principal components can achieve around 100% and 200% improvement for East Asian and European populations, respectively. Conclusion With the optimal parameter confguration on both PCA and random forest, our proposed method can infer the individual genetic ancestry more accurately, given their variants.

RESEARCH ON MILLIMETER WAVE RADAR SIMULATION MODEL FOR INTELLIGENT VEHICLE

Xin Li,Weiwen Deng,Sumin Zhang,Yaxin Li,Shiping Song,Shanshan Wang,Guanyu Wang 한국자동차공학회 2020 International journal of automotive technology Vol.21 No.2

Radar simulation models can effectively overcome the drawbacks of real vehicle experiment and speed up the development process of intelligent vehicle technologies based on millimeter wave radar via virtual testing. However, there are still many gaps between the radar model using in the virtual driving environment and the real radar. In this paper, a novel simulation model of intelligent vehicle millimeter wave radar is proposed. Based on the analysis of the real radar performance in typical application scenes, the radar model considers the mechanism and characteristics of the vehicle radar synthetically and a systematic radar modeling architecture with innovation is introduced. The highlights of this radar model include the design of the RCS simulation model for radar targets with both high accuracy and real-time performance, the establishment of the quantitative false alarm model, missed detection model and measurement error simulation model. Vast amounts of data collected by real vehicle radar are applied to fetch model parameters and verify the accuracy of the radar model. Simulation results show that the proposed model can reach both high reliability and computational efficiency.

AuNP-Loaded Electrospinning Membrane Cooperated with CDs for Periodontal Tissue Engineering

Wang Jie,Zhang Hang,Wang Yan,Liu Xiang,Zhu Weiwen,Jiang Fei,Li Size,Liu Laikui 한국조직공학과 재생의학회 2023 조직공학과 재생의학 Vol.20 No.7

BACKGROUND: Guided bone regeneration (GBR) is commonly used to regenerate periodontal tissue. However, the bone inductivity and antibacterial properties of the GBR membranes currently in use are severely limited. This issue can be resolved by loading growth factors and antibiotics. Bioactive substitutes, such as Au nanoparticles (AuNPs) and carbon quantum dots (CDs), were proposed to prevent the denaturation of osteogenic growth factors and the induction of antibacterial drug resistance. METHODS: Ornidazole was initially used as the raw material to prepare the CDs, followed by the incorporation of an optimal ratio of nanoparticles to produce the electrospun membrane doped with AuNPs and novel traceable antibacterial CDs. The morphology of the membrane was characterized. The adhesion, proliferation, and osteogenic differentiation of cells on the membrane were evaluated in vitro. The antimicrobial characteristics of the membrane were also investigated. The electrospun membrane was implanted into a rat skull defect model in vivo to investigate its osteogenic potential. RESULTS: The blending of nanomaterials did not affect the micro morphology of the fiber, resulting in enhanced mechanical properties. Membranes doped with AuNPs and CDs exhibited excellent biocompatibility, increased ALP activity, improved calcified nodules, and increased expression of osteogenic-associated proteins, in addition to pronounced antibacterial effects. The membrane also demonstrated excellent osteogenic characteristics in rat models. CONCLUSION: The synergistic effect of loaded AuNPs electrospun fiber membrane with CDs can promote periodontal bone regeneration and exert antibacterial activity.

Compressive resistance behavior of UHPFRC encased steel composite stub column

Zhenyu Huang,Xinxiong Huang,Weiwen Li,Jiasheng Zhang 국제구조공학회 2020 Steel and Composite Structures, An International J Vol.37 No.2

To explore the feasibility of eliminating the longitudinal rebars and stirrups by using ultra-high-performance fiber reinforcement concrete (UHPFRC) in concrete encased steel composite stub column, compressive behavior of UHPFRC encased steel stub column has been experimentally investigated. Effect of concrete types (normal strength concrete, high strength concrete and UHPFRC), fiber fractions, and transverse reinforcement ratio on failure mode, ductility behavior and axial compressive resistance of composite columns have been quantified through axial compression tests. The experimental results show that concrete encased composite columns with NSC and HSC exhibit concrete crushing and spalling failure, respectively, while composite columns using UHPFRC exhibit concrete spitting and no concrete spalling is observed after failure. The incorporation of steel fiber as micro reinforcement significantly improves the concrete toughness, restrains the crack propagation and thus avoids the concrete spalling. No evidence of local buckling of rebars or yielding of stirrups has been detected in composite columns using UHPFRC. Steel fibers improve the bond strength between the concrete and, rebars and core shaped steel which contribute to the improvement of confining pressure on concrete. Three prediction models in Eurocode 4, AISC 360 and JGJ 138 and a proposed toughness index (T.I.) are employed to evaluate the compressive resistance and post peak ductility of the composite columns. It is found that all these three models predict close the compressive resistance of UHPFRC encased composite columns with/without the transverse reinforcement. UHPFRC encased composite columns can achieve a comparable level of ductility with the reinforced concrete (RC) columns using normal strength concrete. In terms of compressive resistance behavior, the feasibility of UHPFRC encased steel composite stub columns with lesser longitudinal reinforcement and stirrups has been verified in this study.

Enhanced proton conductivity of sulfonated poly(ether ether ketone) membranes at elevated temperature by incorporating (3-aminopropyl)triethoxysilane-grafted graphene oxide

Shuguo Qu,Chenchen Zhang,Minhui Li,Yan Zhang,Lunbo Chen,Yushuai Yang,Bo Kang,Yiwei Wang,Jihai Duan,Weiwen Wang 한국화학공학회 2019 Korean Journal of Chemical Engineering Vol.36 No.12

Making inexpensive proton exchange membrane with high proton conductivity for the proton exchange membrane fuel cell (PEMFC) is still a challenging problem. Graphene oxide (GO) nanoparticles grafted with (3-aminopropyl) triethoxy silane (APTES) were prepared and then incorporated into sulfonated poly(ether ether ketone) (SPEEK) matrix by solution casting to make the composite proton exchange membrane. The obtained nanoparticles and composite membranes were characterized by XRD, FT-IR, Raman, TGA, SEM, and UTM. GO treated with the silane coupling agent improved the dispersion stability and compatibility of GO in SPEEK, which decreased the agglomeration of GO nanoparticles in the SPEEK membrane. The prepared nanocomposite membranes exhibited better water retention properties and proton conductivity. The proton conductivity of the SPEEK membrane with 2wt% amine functionalized GO (AGO) reached 11.32mS/cm at 120oC, which was 2.45-times higher than that of the pristine SPEEK membrane. The reason was that AGO nanoparticles disperse uniformly in the SPEEK membranes, which provides new channels for proton transfer. The potential application of this composite membrane in the PEMFC was indicated.

p53 overexpression represses androgen-mediated induction ofNKX3.1 in a prostate cancer cell line

Anli Jiang,Chunxiao Yu,Pengju Zhang,Weiwen Chen,Wenwen Liu,Xiaoyan Hu,Jianye Zhang 생화학분자생물학회 2006 Experimental and molecular medicine Vol.38 No.6

Prostate cancer is a disease involving complicated multiple-gene alterations. Both NKX3.1 and p53 are related to prostate cancer and play crucial roles in prostate cancer progression. However, little is known about the relationships and interactions between p53 and NKX3.1 in prostate cancer. We found that NKX3.1 expression is down-regulated by over-expression of wild type (wt) p53 in prostate cancer LNCaP cells. NKX3.1 is down-regulated at both the mRNA and protein levels by p53 overexpression due to either transient transfection of exogenous p53 or induction of endogenous p53. p53 over-expression represses androgen-induced transactivation of NKX3.1 by inhibiting the promoter of the androgen acceptor (AR) gene and by blocking AR-DNA binding activity. In addition, transfection with the p21 expression vector (pPSA-p21) showed that p21 does not reduce NKX3.1 expression, indicating that NKX3.1 expression is not the result of nonspecific effects of cell growth arrest. Our results provide biochemical and cellular biologic evidence that NKX3.1 is down-regulated by p53 over-ex - pression in prostate cancer cells.