- Statistical Methods and Inference
- COVID-19 epidemiological studies
- Genetic Mapping and Diversity in Plants and Animals
- Random Matrices and Applications
- Remote Sensing in Agriculture
- Statistical Methods and Bayesian Inference
- Smart Agriculture and AI
- Advanced Statistical Methods and Models
- Plant Molecular Biology Research
- Leaf Properties and Growth Measurement
- Gene expression and cancer classification
- Spectroscopy and Chemometric Analyses
- Greenhouse Technology and Climate Control
- Time Series Analysis and Forecasting
- Bayesian Methods and Mixture Models
- COVID-19 Pandemic Impacts
- SARS-CoV-2 and COVID-19 Research
- Genetics and Plant Breeding
- Vaccine Coverage and Hesitancy
- Blind Source Separation Techniques
- Functional Brain Connectivity Studies
- Genomics and Phylogenetic Studies
- Advanced Causal Inference Techniques
- Remote Sensing and LiDAR Applications
- Lipid metabolism and biosynthesis
Peking University
2018-2025
Iowa State University
2012-2024
King University
2018-2024
University of Surrey
2020
Instituto de Ciencias Agrarias
2020
University of Nebraska–Lincoln
2017-2018
Identifying interspecies changes in gene regulation, one of the two primary sources phenotypic variation, is challenging on a genome-wide scale. The use paired time-course data cold-responsive expression maize (Zea mays) and sorghum (Sorghum bicolor) allowed us to identify differentially regulated orthologs. While majority transcriptional regulation conserved pairs species specific, initial responses cold appear be more than later responses. In maize, promoters genes with tend contain...
Image-based plant phenotyping facilitates the extraction of traits noninvasively by analyzing large number plants in a relatively short period time. It has potential to compute advanced phenotypes considering whole as single object (holistic phenotypes) or individual components, i.e., leaves and stem (component phenotypes), investigate biophysical characteristics plants. The emergence timing, total present at any point time growth during vegetative stage life cycle maize are significant...
Abstract By proposing a varying coefficient Susceptible-Infected-Removal model (vSIR), we track the epidemic of COVID-19 in 30 provinces China and 15 cities Hubei province, epicenter outbreak. It is found that spread has been significantly slowing down within two weeks from January 27 to February 10th with 87.0% 84.3% reductions reproduction number R 0 among cities, respectively. This suggests extreme control measures implemented since 23, which include cutting off Wuhan many other towns,...
Abstract High‐throughput phenotyping systems provide abundant data for statistical analysis through plant imaging. Before usable can be obtained, image processing must take place. In this study, we used supervised learning methods to segment plants from the background in such images and compared them with commonly thresholding methods. Because obtaining accurate training is a major obstacle using segmentation, novel approach producing labels was developed. We demonstrated that, careful...
Abstract Background Maize (Zea mays ssp. mays) is 1 of 3 crops, along with rice and wheat, responsible for more than one-half all calories consumed around the world. Increasing yield stress tolerance these crops essential to meet growing need food. The cost speed plant phenotyping are currently largest constraints on breeding efforts. Datasets linking new types high-throughput data collected from plants performance same genotypes under agronomic conditions across a wide range environments...
Recent advances have shown that statistical tests for the rank of cross-covariance matrices play an important role in causal discovery. These include partial correlation as special cases and provide further graphical information about latent variables. Existing typically assume all continuous variables can be perfectly measured, yet, practice many only measured after discretization. For example, psychometric studies, level certain personality dimensions a person being discretized into...
ABSTRACT Covariance is an important measure of marginal dependence among variables. However, heterogeneity in subject covariances and regression models for high-dimensional covariance matrices not well studied. Compared to analysis conditional means, modeling much more challenging due the large set free parameters intrinsic positive-definite property that puts constraints on parameters. In this paper, we propose a regularized estimation method coefficients under sufficient necessary positive...
We consider the statistical inference for high-dimensional precision matrices. Specifically, we propose a data-driven procedure constructing class of simultaneous confidence regions subset entries large matrix. The can be applied to test specific structures matrix, and recover its nonzero components. first construct an estimator matrix via penalized node-wise regression. then develop Gaussian approximation approximate distribution maximum difference between estimated true coefficients.A...
We propose a varying coefficient Susceptible-Infected-Removal (vSIR) model that allows changing infection and removal rates for the latest corona virus (COVID-19) outbreak in China. The vSIR together with proposed estimation procedures allow one to track reproductivity of COVID-19 through time assess effectiveness control measures implemented since Jan 23 2020 when city Wuhan was lockdown followed by an extremely high level self-isolation population. Our study finds had been significantly...
Motivated by the latest effort to employ banded matrices estimate a high-dimensional covariance $\Sigma$, we propose test for $\Sigma$ being with possible diverging bandwidth. The is adaptive "large $p$, small $n$" situations without assuming specific parametric distribution data. We also formulate consistent estimator bandwidth of matrix. properties and are investigated theoretical evaluations simulation studies, as well an empirical analysis on protein mass spectroscopy
High-throughput phenotyping system has become more and popular in plant science research. The data analysis for such a typically involves two steps: feature extraction through image processing statistical the extracted features. current approach is to perform those steps on different platforms. We develop package “implant” R both robust functional analysis. For processing, provides methods including thresholding, hidden Markov random field model, morphological operations. analysis, this can...
We study epidemiological characteristics of 25 early COVID-19 outbreak countries, which emphasizes on the reproduction infection and effects government control measures. The is based a vSIADR model allows asymptomatic pre-diagnosis infections to reflect clinical realities, linear mixed-effect analyse association between each country's measures effective number Rt . It finds significant higher stringency in lowering reproduction, shortening effect time epidemic turning point by applying...
Linking natural genetic variation to trait can help determine the functional roles ofdifferent genes. Variations of one or several traits are often assessed separately. High-throughput phenotyping and data mining capture dozens hundreds from same individuals. Here, we test association between markers within a gene many simultaneously. This genome–phenome wide study (GPWAS) is both multi-marker multi-trait test. Genes identified using GPWAS with 260 phenotypic in maize were enriched for genes...
High-throughput plant phenotyping-the use of imaging and remote sensing to record growth dynamics-is becoming more widely used. The first step in this process is typically segmentation, which requires a well-labeled training dataset enable accurate segmentation overlapping plants. However, preparing such data both time labor intensive. To solve problem, we propose image processing pipeline using self-supervised sequential convolutional neural network method for in-field phenotyping systems....
Recent advances in automated plant phenotyping have enabled the collection of time series measurements from same plants a wide range traits at different developmental scales. The availability phenotypic datasets has increased interest statistical approaches for comparing patterns change among genotypes and treatment conditions. Two widely used methods modeling growth with are pointwise analysis variance (ANOVA) parametric sigmoidal curve fitting. Pointwise ANOVA yields discontinuous curves,...
ABSTRACT Maize ( Zea mays ssp. ) is one of three crops, along with rice and wheat, responsible for more than 1/2 all calories consumed around the world. Increasing yield stress tolerance these crops essential to meet growing need food. The cost speed plant phenotyping currently largest constraint on breeding efforts. Datasets linking new types high throughput data collected from plants performance same genotypes under agronomic conditions across a wide range environments are developing...
Yumou Qiu & Song Xi ChenYumou is Assistant Professor, Department of Statistics, University Nebraska-Lincoln, NE 68583-0963 (E-mail: ). Chen Chair Business Statistics and Econometrics, Guanghua School Management Center for Statistical Science, Peking University, Beijing 100651, China, Iowa State Ames, IA 50011-1210 ).We thank the editors, AE three anonymous referees constructive comments suggestions which have improved presentation article. We also Feng Yi Professor Hui Zou sharing code their...
We consider matching problems where the goal is to determine whether two observations randomly drawn from a population with multiple (sub)groups are same (sub)group. This key question in forensic science, items unidentified origins suspects and crime scenes compared objects known set of sources see if they originated source. derive optimal rule under density functions data that minimizes decision error probabilities. Empirically, proposed computed by plugging parametrically estimated using...
Abstract Motivation Large-scale gene expression studies allow network construction to uncover associations among genes. To study direct genes, partial correlation-based networks are preferred over marginal correlations. However, FDR control for is not well-studied. In addition, currently available methods cannot take existing biological knowledge help while controlling FDR. Results this paper, we propose a method called Partial Correlation Graph with Information Incorporation (PCGII). PCGII...
Incorporating the auxiliary information into survey estimation is a fundamental problem in sampling. Calibration weighting popular tool for incorporating information. The calibration method of Deville and Sarndal (1992) uses distance measure between design weights final to solve optimization with constraints. This paper introduces novel framework that leverages generalized entropy as objective function optimization, where play role constraints ensure consistency, rather than being part...
Abstract Chilling stress threatens plant growth and development, particularly affecting membrane fluidity cellular integrity. Understanding responses to chilling is important for unraveling the molecular mechanisms of tolerance. Whereas core transcriptional tolerance are conserved across species, associated changes in lipids appear be less conserved, as which affected by varies species. Here, we investigated gene expression response during one 24 h cycle chilling-tolerant foxtail millet...