A case study of competing risk analysis in the presence of missing data

Limei Zhou,Peter C. Austin,Abdel-Qadir 한국통계학회 2023 Communications for statistical applications and me Vol.30 No.1

Observational data with missing or incomplete data are common in biomedical research. Multiple imputation is an effective approach to handle missing data with the ability to decrease bias while increasing statistical power and efficiency. In recent years propensity score (PS) matching has been increasingly used in observational studies to estimate treatment effect as it can reduce confounding due to measured baseline covariates. In this paper, we describe in detail approaches to competing risk analysis in the setting of incomplete observational data when using PS matching. First, we used multiple imputation to impute several missing variables simultaneously, then conducted propensity-score matching to match statin-exposed patients with those unexposed. Afterwards, we assessed the effect of statin exposure on the risk of heart failure-related hospitalizations or emergency visits by estimating both relative and absolute effects. Collectively, we provided a general methodological framework to assess treatment effect in incomplete observational data. In addition, we presented a practical approach to produce overall cumulative incidence function (CIF) based on estimates from multiple imputed and PS-matched samples.

Simultaneous Estimation of Low- and High-Order Functional Connectivity for Identifying Mild Cognitive Impairment

Zhou, Yueying,Qiao, Lishan,Li, Weikai,Zhang, Limei,Shen, Dinggang Frontiers Media S.A. 2018 Frontiers in neuroinformatics Vol.12 No.-

<P>Functional connectivity (FC) network has been becoming an increasingly useful tool for understanding the cerebral working mechanism and mining sensitive biomarkers for neural/mental disease diagnosis. Currently, Pearson's Correlation (PC) is the simplest and most commonly used scheme in FC estimation. Despite its empirical effectiveness, PC only encodes the low-order (i.e., second-order) statistics by calculating the pairwise correlations between network nodes (brain regions), which fails to capture the high-order information involved in FC (e.g., the correlations among different edges in a network). To address this issue, we propose a novel FC estimation method based on Matrix Variate Normal Distribution (MVND), which can capture both low- and high-order correlations simultaneously with a clear mathematical interpretability. Specifically, we first generate a set of BOLD subseries by the sliding window scheme, and for each subseries we construct a temporal FC network by PC. Then, we employ the constructed FC networks as samples to estimate the final low- and high-order FC networks by maximizing the likelihood of MVND. To illustrate the effectiveness of the proposed method, we conduct experiments to identify subjects with Mild Cognitive Impairment (MCI) from Normal Controls (NCs). Experimental results show that the fusion of low- and high-order FCs can generally help to improve the final classification performance, even though the high-order FC may contain less discriminative information than its low-order counterpart. Importantly, the proposed method for simultaneous estimation of low- and high-order FCs can achieve better classification performance than the two baseline methods, i.e., the original PC method and a recent high-order FC estimation method.</P>

Three-Month Follow-Up Study of Survivors of Coronavirus Disease 2019 after Discharge

Liang Limei,Yang Bohan,Jiang Nanchuan,Fu Wei,He Xinliang,Zhou Yaya,Ma Wan-Li,Wang Xiaorong 대한의학회 2020 Journal of Korean medical science Vol.35 No.47

Background: Most patients including health care workers (HCWs) survived the coronavirus disease 2019 (COVID-19), however, knowledge about the sequelae of COVID-19 after discharge remains limited. Methods: A prospectively observational 3-month follow-up study evaluated symptoms, dynamic changes of severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) IgG and IgM, lung function, and high resolution computed tomography (HRCT) of survivors of COVID-19 after discharge at Wuhan Union Hospital, China. Results: Seventy-six survivors (55 females) with a mean age of 41.3 ± 13.8 years were enrolled, and 65 (86%) were HCWs. A total of 69 (91%) patients had returned to their original work at 3-months after discharge. Most of the survivors had symptoms including fever, sputum production, fatigue, diarrhea, dyspnea, cough, chest tightness on exertion and palpitations in the three months after discharge. The serum troponin-I levels during the acute illness showed high correlation with the symptom of fatigue after hospital discharge (r = 0.782; P = 0.008) and lymphopenia was correlated with the symptoms of chest tightness and palpitations on exertion of patients after hospital discharge (r = −0.285, P = 0.027; r = −0.363, P = 0.004, respectively). The mean values of forced expiratory volume in 1 second (FEV1), forced vital capacity (FVC), FEV1/FVC, total lung capacity and diffusion capacity were all normal (> 80% predicted) and lung HRCTs returned to normal in most of the patients (82%), however, 42% of survivors had mild pulmonary function abnormalities at 3-months after discharge. SARS-CoV-2 IgG turned negative in 11% (6 of 57 patients), 8% (4 of 52 patients) and 13% (7 of 55 patients), and SARS-CoV-2 IgM turned negative in 72% (41 of 57 patients), 85% (44 of 52 patients) and 87% (48 of 55 patients) at 1-month, 2-months and 3-months after discharge, respectively. Conclusion: Infection by SARS-CoV-2 caused some mild impairments of survivors within the first three months of their discharge and the duration of SARS-CoV-2 antibody was limited, which indicates the necessity of long-term follow-up of survivors of COVID-19.

One-Pot Hydrothermal Synthesis of Sulfur-Doped SnO2 Nanoparticles and their Enhanced Photocatalytic Properties

Lin Ma,Limei Xu,Xuyao Xu,Xiaoping Zhou,Lingling Zhang 성균관대학교(자연과학캠퍼스) 성균나노과학기술원 2016 NANO Vol.11 No.3

Sulfur-doped SnO2 nanoparticles with ultrafine sizes have been successfully prepared by a one-pot hydrothermal method. The obtained samples are characterized by X-ray diffraction (XRD), energy dispersive X-ray spectroscopy (EDX), X-ray photoelectron spectroscopy (XPS), high resolution transmission electron microscopy (HRTEM), thermogravimetric (TG), analyzer UV-Vis spectroscopy, photoluminescence (PL) and electrochemical impedance spectroscopy (EIS). The experimental results indicate that the doping level of sulfur element as well as the bandgaps of SnO2 can be controlled to a certain extent by varying the amount of L-cysteine (L-cys). When evaluated as photocatalysts in the degradation of rhodamine B (RhB) and reduction of Cr(VI) under visible light region, the resultant sulfur-doped SnO2 nanoparticles demonstrate obviously enhanced photocatalytic activities due to the markedly improved visible light response and effective separation of the photo-generated electron–hole pairs.

Topography of the 660km Discontinuity Beneath the Northeast China

(Shaoxian Zang),(Limei Zhou),(Youlin Chen) 한국지진연구소 1996 MODERN SEISMOLOGY Vol.- No.-

Microwave-Assisted Hydrothermal Preparation of SnO2/MoS2 Composites and Their Electrochemical Performance

Lin Ma,Xiaoping Zhou,Limei Xu,Xuyao Xu,Lingling Zhang 성균관대학교(자연과학캠퍼스) 성균나노과학기술원 2016 NANO Vol.11 No.2

We introduce a two-step hydrothermal and microwave method to prepare novel SnO2/MoS2 composites. The as-prepared samples are well characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM) and high resolution transmission electron microscopy (HRTEM). The experimental results indicate that the SnO2/MoS2 composites are composed of MoS2 nanosheets and ultrafine SnO2 nanoparticles with mean size of 3–4 nm which are well-distributed and anchored on the surface of MoS2 nanosheets. The resultant composites demonstrate prominently improved electrochemical performances, which could be attributed to the unique and robust microstructures and synergetic effect between MoS2 and SnO2.

Enhanced anticorrosion of methyl acrylate by covalent bonded N-alkylpyridinium bromide for X70 steel in 5 M HCl

Guangqiang Xia,Xiaohui Jiang,Limei Zhou,Yinwen Liao,Ming Duan,Hu Wang,Qiang Pu,Jie Zhou 한국공업화학회 2015 Journal of Industrial and Engineering Chemistry Vol.27 No.-

Methyl acrylate connects covalently with alkylpyridinium bromide to build N-alkyl-3-(2-methox-ycarbonyl-vinyl)pyridinium bromide (MPA-n, n = 8–14). MPA-n inhibited X70 steel from corrosion in5 M HCl with inhibition efficiency of 98% and the inhibition capacity was in the order of MPA-14 > MPA-12 > MPA-10 > MPA-8. Quantum chemical calculation unveiled that MPA-n adsorbed on iron surface byO atom in MA moiety offering electrons to d orbital of iron and by p* orbital of pyridinium acceptingelectrons from iron, and MPA-n with longer alkylchain possessed higher EHOMO, lower DE, largermolecular volume and dipole moment, all suggesting the same corrosion inhibition order as that inexperiments

A repressor motif-containing poplar R3 MYB-like transcription factor regulates epidermal cell fate determination and anthocyanin biosynthesis in Arabidopsis

Qingnan Hu,Li Yang,Shanda Liu,Limei Zhou,Xutong Wang,Wei Wang,Ling Cai,Xiuju Wu,Ying Chang,Shucai Wang 한국식물학회 2016 Journal of Plant Biology Vol.59 No.5

Single-repeat R3 MYB transcription factors (R3 MYBs) regulate epidermal cell fate determination in Arabidopsis through a lateral inhibition mechanism. Previously we have shown that poplar R3 MYB genes regulate trichome formation when expressed in Arabidopsis. Here we report the identification and functional characterization of a poplar R3 MYB-like protein, Populus trichocarpa R3 MYB-LIKE1 (PtrRML1). So far all the MYB transcription factors identified have a highly conserved N-terminal DNA-binding domain composed of MYB repeats and a highly variable C-terminal domain. However, PtrRML1 has a single R3 MYB repeat at its C-terminal and an LxLxL repressor motif-containing N-terminal, and with amino acids about twice of that of the known R3 MYBs. PtrRML1 is localized in nuclear in transgenic Arabidopsis plants expressing PtrRML1-GFP. PtrRML1 repressed reporter gene expression in protoplasts, and it interacted with GL3 in plant cells. Expression of PtrRML1 in Arabidopsis resulted in glabrous phenotypes, increased number of root hairs, and decreased anthocyanin accumulation. Consist with these observation, the expression levels of some MBW component genes and some of their target genes involved in the regulation of epidermal cell fate determination and anthocyanin biosynthesis, including TRY, CPC, ETC1, GL1, GL2, DFR, UF3GT and TT8 were reduced in the transgenic plants.

Epigenetic inactivation of ACAT1 promotes epithelial-mesenchymal transition of clear cell renal cell carcinoma

Han Peipei,Wu Shu,Li Limei,Li Danping,Zhao Jun,Zhang Haishan,Wang Yifang,Zhong Xuemin,Zhang Zhe,Li Ping,Matskova Liudmila,Zhou Xiaoying 한국유전학회 2022 Genes & Genomics Vol.44 No.4

Background: Acetyl-CoA acyltransferase 1 (ACAT1) is a key enzyme catalyzing the production of mitochondrial ketone bodies. We have shown that ACAT1 is down-regulated in kidney renal clear cell carcinoma (KIRC) previously. Objective: To investigate the reasons for downregulation of ACAT1 in KIRC and explore the underlying mechanisms involved in metastatic inhibition regulated by ACAT1. Methods: The Gene Expression Omnibus (GEO) database was queried for meta-analysis of ACAT1 mRNA expression in KIRC. The UALCAN website was used to compare the methylation levels of the ACAT1 promoter region in KIRC and normal tissues. RT-qPCR was used to quantitate ACAT1 transcription levels. The GCBI and Tarbase V.8 databases were used to predict miRNAs that may target the mRNA of ACAT1. The correlation between mRNA expression of ACAT1, MMP7 (matrix metallopeptidase 7), CDH1 (E-cadherin), EpCAM (epithelial cell adhesion molecule), and VIM (vimentin) was analyzed. Extracellular MMP7 protein was quantitated using an ELISA assay. Results: The methylation level of the ACAT1 promoter region in KIRC was significantly higher than that in the normal kidney tissues. The ACAT1 mRNA expression in the KIRC cell lines was restored after treatment with 5-aza-dC (p < 0.05). MiR-21-5p is a conserved microRNA targeting ACAT1. It is expressed at a significantly higher level in KIRC than in normal tissues (p < 0.001). MiR-21-5p miRNA expression negatively correlates with ACAT1 mRNA expression. The expression of miR-21-5p is higher at the T3-T4 stages and in the histologic grades G3-G4. Patients with high miR-21-5p expression tended to have lower overall survival, suggesting that miR-21-5p could serve as a potentially valuable diagnostic biomarker for KIRC (AUC = 0.957; p < 0.001). A mimetic of miR-21-5p inhibited the expression of ACAT1 mRNA and protein. In addition, ACAT1 mRNA expression positively correlates with CDH1 and EpCAM but is negatively correlated with VIM. Overexpression of ACAT1 suppresses the secretion of MMP7 in KIRC cells. Conclusion: Expression of ACAT1 in KIRC is controlled at two levels, firstly by the hypermethylation of the ACAT1 promoter region and secondly by overexpression of miR-21-5p. Downregulation of ACAT1 expression correlates with epithelial-mesenchymal transition (EMT).

Inactivation of epithelial sodium ion channel molecules serves as effective diagnostic biomarkers in clear cell renal cell carcinoma

Zheng Qian,Wang Yifang,Zhao Ran,Han Peipei,Zhao Jun,Li Limei,Zhou Xiaohui,Li Ping,Mo Yingxi,Pan Xinli,Luo Wenqi,Zhou Xiaoying 한국유전학회 2023 Genes & Genomics Vol.45 No.7

Background Non-voltage-gated sodium channel, also known as the epithelial sodium channel (ENaC), formed by heteromeric complexes consisting of SCNN1A, SCNN1B, and SCNN1G, is responsible for maintaining sodium ion and body fluid homeostasis in epithelial cells. However, no systematic study of SCNN1 family members has been conducted in renal clear cell carcinoma (ccRCC) to date. Objective To investigate the abnormal expression of SCNN1 family in ccRCC and its potential correlation with clinical parameters. Methods The transcription and protein expression levels of SCNN1 family members in ccRCC were analyzed based on the TCGA database, and were confirmed by quantitative RT-PCR and immunohistochemical staining assays, respectively. The area under curve (AUC) was used to evaluate the diagnostic value of SCNN1 family members for ccRCC patients. Results The mRNA and protein expression of SCNN1 family members was significantly downregulated in ccRCC compared with normal kidney tissues, which might be due to DNA hypermethylation in the promoter region. It is worth noting that the AUC of SCNN1A, SCNN1B, and SCNN1G were 0.965, 0.979, and 0.988 based on the TCGA database (p < 0.0001), respectively. The diagnostic value was even higher when combing these three members together (AUC = 0.997, p < 0.0001). Intriguingly, the mRNA level of SCNN1A was significantly lower in females compared with males, while SCNN1B and SCNN1G were increased with the progression of ccRCC and remarkably associated with a worse outcome for patients. Conclusion The aberrantly decrease of SCNN1 family members might serve as valuable biomarkers for the diagnosis of ccRCC.