A5052889294- Stochastic processes and financial applications
- Bayesian Methods and Mixture Models
- Risk and Portfolio Optimization
- Markov Chains and Monte Carlo Methods
- Statistical Methods and Inference
- Distributed Sensor Networks and Detection Algorithms
- Auction Theory and Applications
- Economic theories and models
- Metaheuristic Optimization Algorithms Research
- Stochastic processes and statistical mechanics
- Advanced Algorithms and Applications
- Random Matrices and Applications
- Genetic and phenotypic traits in livestock
- Genetic Associations and Epidemiology
- Advanced Multi-Objective Optimization Algorithms
- Data-Driven Disease Surveillance
- Bioinformatics and Genomic Networks
- Information Retrieval and Data Mining
- Influenza Virus Research Studies
- Bayesian Modeling and Causal Inference
- Advanced Statistical Process Monitoring
- Optimization and Search Problems
- COVID-19 epidemiological studies
- Financial Risk and Volatility Modeling
- Mathematical Biology Tumor Growth
Yangtze University
2021-2025
Texas A&M University
2014-2024
Qingdao University of Science and Technology
2024
University of Bristol
2023
Guizhou Electric Power Design and Research Institute
2019
Rice University
2005-2018
Silesian University of Technology
2018
Concordia University
2017
Baylor College of Medicine
2016-2017
Xichang University
2014-2016
DNA methylation is an epigenetic mechanism central to development and maintenance of complex mammalian tissues, but our understanding its role in intestinal limited.We use whole genome bisulfite sequencing, find that differentiation mouse colonic stem cells epithelium not associated with major changes methylation. However, we detect extensive dynamic their progeny during the suckling period, suggesting postnatal this cell population. We increases at 3' CpG islands (CGIs) correlate...
Abstract Recent models propose deoxyribonucleic acid methylation of key neuro-regulatory genes as a molecular mechanism underlying the increased risk mental disorder associated with early life adversity (ELA). The goal this study was to examine association ELA oxytocin receptor gene ( OXTR) among young adults. Drawing from 21-year longitudinal cohort, we compared adulthood OXTR frequency 46 adults (23 males and 23 females) selected for high or low exposure based on childhood socioeconomic...
Mexicans are a recent admixture of Amerindians, Europeans, and Africans. We performed local ancestry analysis Mexican samples from two genome-wide association studies obtained dbGaP, discovered that at the MHC region have excessive African ancestral alleles compared to rest genome, which is hallmark selection for admixed samples. The estimated coefficients 0.05 0.07 datasets, put our finding among strongest known selections observed in humans, namely, lactase northern Europeans sickle-cell...
The seminal work of Morgan & Rubin (2012) considers rerandomization for all the units at one time.In practice, however, experimenters may have to rerandomize sequentially. For example, a clinician studying rare disease be unable wait perform an experiment until experimental are recruited. Our offers mathematical framework sequential designs, where enrolled in groups. We formulate adaptive procedure balancing treatment/control assignments over some continuous or binary covariates, using...
Surrogate-assisted evolutionary algorithms (SAEAs), which combine the search capabilities of (EAs) with predictive surrogate models, are effective methods for solving expensive optimization problems (EOPs). However, over-reliance on accuracy model causes performance most SAEAs to decrease drastically increase in dimensionality. To tackle this challenge, paper proposes a surrogate-assisted gray prediction evolution (SAGPE) algorithm based (GPE). In SAGPE, both global and local constructed...
Deep mutational scanning has been used to create high-resolution DNA sequence maps that illustrate the functional consequences of large numbers point mutations. However, this approach not yet applied libraries genes created by random circular permutation, an engineering strategy is open reading frames express proteins with altered contact order. We describe a new method, termed permutation profiling sequencing (CPP-seq), which combines one-step transposon mutagenesis protocol for creating...
We show that under the null, 2 log(Bayes factor) is asymptotically distributed as a weighted sum of chi-squared random variables with shifted mean. This claim holds for Bayesian multi-linear regression family conjugate priors, namely, normal-inverse-gamma prior, g-prior, and normal prior. Our results have three immediate impacts. First, we can compute analytically p-value associated Bayes factor without need permutation. provide software package evaluate efficiently accurately. Second, null...
Abstract Yang et al. proved that the symmetric random walk Metropolis–Hastings algorithm for Bayesian variable selection is rapidly mixing under mild high-dimensional assumptions. We propose a novel Markov chain Monte Carlo (MCMC) sampler using an informed proposal scheme, which we prove achieves much faster time independent of number covariates, assumptions To best our knowledge, this first result rigorously shows rate MCMC methods can be fast enough to offset computational cost local...
Consider the motion of a Brownian particle in three dimensions, whose two spatial coordinates are standard motions with zero drift, and remaining (unknown) coordinate is (known) nonzero drift. Given that position being observed real time, problem to detect as soon possible minimal probabilities wrong terminal decisions, which has We solve this Bayesian formulation, under any prior drift coordinates, when passage time penalised linearly. Finding exact solution including rigorous treatment its...
Bayesian variable selection regression (BVSR) is able to jointly analyze genome-wide genetic datasets, but the slow computation via Markov chain Monte Carlo (MCMC) hampered its wide-spread usage. Here we present a novel iterative method solve special class of linear systems, which can increase speed BVSR model-fitting tenfold. The hinges on complex factorization sum two matrices and solution path resides in domain (instead real domain). Compared Gauss-Seidel method, converges almost...
Abstract The coronavirus disease of 2019 (COVID-19) is a pandemic. To characterize its transmissibility, we propose Bayesian change point detection model using daily actively infectious cases. Our builds on Poisson segmented regression that can 1) capture the epidemiological dynamics under changing conditions caused by external or internal factors; 2) provide uncertainty estimates both number and locations points; 3) adjust any explanatory time-varying covariates. be used to evaluate public...
Suppose you have one unit of stock, currently worth 1, which must sell before time T. The Optional Sampling Theorem tells us that whatever stopping we choose to sell, the expected discounted value get when will be 1. however are able see a units into future, and base our rule on that; should do better than But how much can do? And would exploit additional information? optimal solution this problem never found, but in paper establish remarkably close bounds problem, derive fairly simple...
With the increasing number of mobile users and applications, data services on-demand suggestions gradually play an important role in life. In this work, based on wireless personal communication network, we propose a novel prediction method user traffic density interest real time positioning information. Service provider could send precise information to target ensure quality push service. Then execute simulation using around Tsinghua University obtained from network service provider,...
Structure learning via MCMC sampling is known to be very challenging because of the enormous search space and existence Markov equivalent DAGs. Theoretical results on mixing behavior are lacking. In this work, we prove rapid a random walk Metropolis-Hastings algorithm, which reveals that complexity Bayesian sparse equivalence classes grows only polynomially in $n$ $p$, under some high-dimensional assumptions. A series consistency obtained, including strong selection an empirical Bayes model...
The coronavirus disease of 2019 (COVID-19) is a pandemic. To characterize its transmissibility, we propose Bayesian change point detection model using daily actively infectious cases. Our builds on Poisson segmented regression that 1) capture the epidemiological dynamics under changing conditions caused by external or internal factors; 2) provide uncertainty estimates both number and locations points; 3) has potential to adjust for any time-varying covariate effects. can be used evaluate...
Abstract Recent progress in microdissection and DNA sequencing has facilitated the subsampling of multi-focal cancers organs such as liver several hundred spots, helping to determine pattern mutations each these spots. This led construction genealogies primary, secondary, tertiary, so forth, foci tumor. These studies have diverse conclusions concerning Darwinian (selective) or neutral evolution cancer. Mathematical models development tumors been devised support claims. We offer a model for...
We study the Bayesian multi-task variable selection problem, where goal is to select activated variables for multiple related datasets simultaneously. propose a new variational Bayes algorithm which generalizes and improves recently developed "sum of single effects" model Wang et al. Motivated by differential gene network analysis in biology, we further extend our method joint structure learning directed acyclic graphical models, problem known be computationally highly challenging. novel...
Currently, research on the lattice Boltzmann method mainly focuses its numerical simulation and applications, there is an increasing demand for large-scale simulations in practical scenarios. In response to this situation, study successfully implemented a heterogeneous parallel algorithm using OpenMP, MPI, Pthread, OpenCL technologies “Dongfang” supercomputer system. The accuracy effectiveness of were verified through lid-driven cavity flow simulation. paper focused optimizing four aspects:...
Surrogate-assisted evolutionary algorithms (SAEAs) have been proved to be promising approaches the expensive optimization problems (EOPs). However, balance between convergence speed and effectiveness of SAEAs needs further optimized. To solve this problem, a surrogate-assisted grey prediction evolution algorithm (SAGPE) is proposed in paper. In SAGPE, global local surrogate model are constructed respectively perform searches alternately. Grey employed generate offspring through even...
Abstract Different from the conventional energy distribution network, which solely relies on distribution, novel system gradually exhibits a complex and interconnected form of various resources, such as source network load storage. These resources possess characteristics “multi-point, wide area, small amount” distribution. Investigating potential for wide-area voltage regulation by using distributed like renewable power generation, storage, flexible loads is crucial constructing highly...