Show pageOld revisionsBacklinksBack to top This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong. CKG Edit ====== Winston Haynes ====== {{people:haynes.jpg?200|}} 5th Year PhD Candidate in Biomedical Informatics Previous Education: Hendrix College, BA in Bioinformatics ====Research Interests==== ==Meta-Analysis Methods== We have developed the MetaIntegrator R package (available on [[https://cran.r-project.org/web/packages/MetaIntegrator/index.html|CRAN]]). MetaIntegrator covers the entire gene expression meta-analysis pipeline, including data download from GEO, meta-analysis of the datasets, and visualization of the results ([[https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5167529/|paper]]). We systematically examined best practices in gene expression meta-analysis to further optimize our methodologies ([[http://nar.oxfordjournals.org/content/early/2016/09/14/nar.gkw797.abstract|paper]]). ==Meta-Analysis for Biomedical Researchers== We have developed [[http://metasignature.stanford.edu|MetaSignature]] to enable biological researchers to interact with the results from our meta-analyses of 104 human diseases across 619 studies and 41,338 samples ([[https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5167529/|paper]]). We believe that biological researchers will find this resource immensely valuable because it contains gene expression values derived from our meta-analysis across these diseases, so researchers will be able to electronically validate hypotheses about gene-disease relationships before performing any wet lab experiments. MetaSignature is the first resource to make large-scale gene expression meta-analysis results publicly available and has been designed with a focus on usability for biomedical researchers. ==Disease Similarity== As our molecular understanding of and treatments for human diseases have improved, researchers seek to expand these findings to other related diseases based on the existing literature. We integrated data on gene expression, genetic variation, electronic health records, ontologies, drug indications, and publications to paint a complete picture of human disease. Our methods illustrate how such diverse data sources can be integrated to provide an unbiased understanding of disease which connects molecular understanding with clinical practice. ==Research Bias== A thorough understanding of disease relationships requires a rigorous and unbiased understanding of each disease. Unfortunately, our analysis demonstrated a prominent bias in disease research towards well-characterized genes instead of those with the strongest molecular implications. We hypothesize that researchers overlook genes that are highly differentially expressed in disease because they are not well annotated. ==Pathway Analysis== Differential Expression Analysis for Pathways ([[http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1002967|DEAP]]) is a novel method to characterize differential expression in biological pathways based on the most differentially expressed subpath. Focusing on the most differentially expressed subpath led to a significant increase in statistical power in comparison to existing approaches. ====Publications==== [[https://scholar.google.com/citations?user=BRk6-IgAAAAJ&hl=en|My Google Scholar Profile]] {{rss>http://www.feed43.com/winngooglescholar.xml 100 description 12h}} Winn Haynes people/winstonhaynes.txt Last modified: 2017/12/05 13:33by hayneswa