A5051461953- Gene expression and cancer classification
- Statistical Methods in Clinical Trials
- Bioinformatics and Genomic Networks
- Optimal Experimental Design Methods
- Carcinogens and Genotoxicity Assessment
- Molecular Biology Techniques and Applications
- Effects and risks of endocrine disrupting chemicals
- Statistical Methods and Bayesian Inference
- Computational Drug Discovery Methods
- Statistical Methods and Inference
- Toxic Organic Pollutants Impact
- DNA Repair Mechanisms
- Advanced Statistical Process Monitoring
- Bayesian Methods and Mixture Models
- Advanced Statistical Methods and Models
- Salmonella and Campylobacter epidemiology
- Cancer-related Molecular Pathways
- Pharmacogenetics and Drug Metabolism
- Genomics and Chromatin Dynamics
- Machine Learning in Bioinformatics
- Data Mining Algorithms and Applications
- Biomedical Text Mining and Ontologies
- Metabolomics and Mass Spectrometry Studies
- Pesticide Residue Analysis and Safety
- Listeria monocytogenes in Food Safety
University of Louisville
2022
United States Food and Drug Administration
2010-2021
National Center for Toxicological Research
2010-2021
University of Arkansas for Medical Sciences
2017-2018
China Medical University
2007-2016
Food and Drug Administration
2015
Zimmer Biomet (United States)
2004-2013
United States Department of Health and Human Services
2009
Center for Food Safety and Applied Nutrition
2008
Institute of Statistical Science, Academia Sinica
2007
Topic modelling is an active research field in machine learning. While mainly used to build models from unstructured textual data, it offers effective means of data mining where samples represent documents, and different biological endpoints or omics words. Latent Dirichlet Allocation (LDA) the most commonly topic method across a wide number technical fields. However, model development can be arduous tedious, requires burdensome systematic sensitivity studies order find best set parameters....
Abstract Background The acceptance of microarray technology in regulatory decision-making is being challenged by the existence various platforms and data analysis methods. A recent report (E. Marshall, Science , 306, 630–631, 2004), extensively citing study Tan et al . ( Nucleic Acids Res ., 31, 5676–5684, 2003), portrays a disturbingly negative picture cross-platform comparability, and, hence, reliability technology. Results We reanalyzed Tan's dataset found that intra-platform consistency...
ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTRole of interfacial phenomena in the behavior alumina-supported palladium crystallites oxygenJames J. Chen and Eli RuckensteinCite this: Phys. Chem. 1981, 85, 11, 1606–1612Publication Date (Print):May 1, 1981Publication History Published online1 May 2002Published inissue 1 1981https://pubs.acs.org/doi/10.1021/j150611a029https://doi.org/10.1021/j150611a029research-articleACS PublicationsRequest reuse permissionsArticle...
Both adolescent substance use and depression are major public health problems, have the tendency to co-occur. Thousands of articles on or been published. It is labor intensive time consuming extract huge amounts information from cumulated collections. Topic modeling offers a computational tool find relevant topics by capturing meaningful structure among collections documents.In this study, total 17,723 abstracts PubMed published 2000 2014 were downloaded as objects, Latent Dirichlet...
Abstract Assessment of therapeutic equivalence or non‐inferiority between two medical diagnostic procedures often involves comparisons the response rates paired binary endpoints. The commonly used and accepted approach to assessing is by comparing asymptotic confidence interval on difference with some clinical meaningful limits. This paper investigates test statistics, a Wald‐type (sample‐based) statistic restricted maximum likelihood estimation (RMLE‐based) statistic, assess based sample...
Testing for significance with gene expression data from DNA microarray experiments involves simultaneous comparisons of hundreds or thousands genes. If R denotes the number rejections (declared significant genes) and V false rejections, then V/R, if > 0, is proportion rejected hypotheses. This paper proposes a model distribution conditional given R, / R. Under independence assumption, convolution two binomials has noncentral hypergeometric distribution. an equicorrelated model, distributions...
Many researchers are concerned with the comparability and reliability of microarray gene expression data. Recent completion MicroArray Quality Control (MAQC) project provides a unique opportunity to assess reproducibility across multiple sites platforms. The MAQC analysis presented for conclusion inter- intra-platform comparability/reproducibility measurements is inadequate. We evaluate reproducibility/comparability data 12901 common genes in four titration samples generated from five...
Microarray-based measurement of mRNA abundance assumes a linear relationship between the fluorescence intensity and dye concentration. In reality, however, calibration curve can be nonlinear.By scanning microarray scanner slide containing known concentrations fluorescent dyes under 18 PMT gains, we were able to evaluate differences in characteristics Cy5 Cy3. First, for same gain is nonlinear at both high low ends. Second, degree nonlinearity depends on gain. Third, two PMTs (for Cy3) behave...
Abstract This article presents a quantitative procedure for using "benchmark dose" to obtain low-dose risk estimates reproductive and developmental toxic effects. combines the best features of previously proposed methods handling litter effects teratology data currently used assessment. The beta-binomial distribution is account effects, Weibull dose—response model modeling teratogenic A benchmark dose, defined be lowest dose at which excess does not exceed 1% with 95% confidence, replace...
Abstract Motivation: Gene class testing (GCT) or gene set analysis (GSA) is a statistical approach to determine whether some functionally predefined sets of genes express differently under different experimental conditions. Shortcomings the Fisher's exact test for overrepresentation are illustrated by an example. Most alternative GSA methods developed data collected from two conditions, and most based on univariate gene-by-gene statistic assume independence among in set. A multivariate...
The big data moniker is nowhere better deserved than to describe the ever-increasing prodigiousness and complexity of biological medical datasets. New methods are needed generate test hypotheses, foster interpretation, build validated predictors. Although multivariate techniques such as cluster analysis may allow researchers identify groups, or clusters, related variables, accuracies effectiveness traditional clustering diminish for large hyper dimensional Topic modeling an active research...
Abstract When a large number of statistical tests is performed, the chance false positive findings could increase considerably. The traditional approach to control probability rejecting at least one true null hypothesis, familywise error rate (FWE). To improve power detecting treatment differences, an alternative expected proportion errors among rejected hypotheses, discovery (FDR). some hypotheses are not true, from either FWE- or FDR-controlling procedure usually lower than designed level....
Genome-wide association studies (GWAS) aim to identify genetic variants (usually single nucleotide polymorphisms [SNPs]) across the entire human genome that are associated with phenotypic traits such as disease status and drug response. Highly accurate reproducible genotype calling paramount since errors introduced by algorithms can lead inflation of false associations between phenotype. Most currently used for GWAS based on multiple arrays. Because hundreds gigabytes (GB) raw data generated...
Cancer survival studies are commonly analyzed using survival-time prediction models for cancer prognosis. A number of different performance metrics used to ascertain the concordance between predicted risk score each patient and actual time, but these can sometimes conflict. Alternatively, patients divided into two classes according a threshold, binary classifiers applied predict patient's class. Although this approach has several drawbacks, it does provide natural such as positive negative...
Before conducting a microarray experiment, one important issue that needs to be determined is the number of arrays required in order have adequate power identify differentially expressed genes. This paper discusses some crucial issues problem formulation, parameter specifications, and approaches are commonly proposed for sample size estimation experiments. Common methods formulated as minimum necessary achieve specified sensitivity (proportion detected truly genes) on average at false...
Abstract Motivation: A microarray experiment is a multi-step process, and each step potential source of variation. There are two major sources variation: biological variation technical This study presents variance-components approach to investigating animal-to-animal, between-array, within-array day-to-day variations for data sets. The first set involved estimation variances pooled control treated RNA samples. variance components included nested variances: between-section (the upper-...
A significant limitation to the analytical accuracy and precision of dual-labeled spotted cDNA microarrays is signal error due dye bias. Transcript-dependent bias may be gene-specific differences incorporation two distinctly different chemical dyes resultant differential hybridization efficiencies these chemically targets for same probe. Several approaches were used assess minimize effects on fluorescent signals maximize experimental design efficiency a cell culture experiment. Dye was...
Drug-induced liver injury (DILI) is the primary adverse event that results in withdrawal of drugs from market and a frequent reason for failure drug candidates development. The Liver Toxicity Biomarker Study (LTBS) an innovative approach to investigate DILI because it compares molecular events produced vivo by compound pairs (a) are similar structure mechanism action, (b) associated with few or no signs toxicity preclinical studies, (c) show marked differences hepatotoxic potential. LTBS...