{"title":"Comparison of linear weighting schemes for perfect match and mismatch gene expression levels from microarray data.","authors":"T Mark Beasley, Janet K Holt, David B Allison","doi":"10.2165/00129785-200505030-00006","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Data analytic approaches to Affymetrix microarray data include: (a) a covariate model, in which the observed signal is some estimated linear function of perfect match (PM) and mismatch (MM) signals; (b) a difference model [PM-MM]; and (c) a PM-only model, in which MM data is not utilized.</p><p><strong>Methods: </strong>By decomposing the correlations among the variables in the statistical model and making certain assumptions, we theoretically derive the statistical model that reflects the actual gene expression level under a variety of conditions expected in microarray data.</p><p><strong>Results and conclusion: </strong>When modeling non-systematic variation, the covariate model provides maximum flexibility and often reflects the actual gene expression levels better than the difference model. However, the PM-only model demonstrates superior power in an overwhelming majority of realistic situations, which provides theoretical support for the current trend to employ PM-only models in microarray data analyzes.</p>","PeriodicalId":72171,"journal":{"name":"American journal of pharmacogenomics : genomics-related research in drug development and clinical practice","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2005-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.2165/00129785-200505030-00006","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"American journal of pharmacogenomics : genomics-related research in drug development and clinical practice","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.2165/00129785-200505030-00006","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
Background: Data analytic approaches to Affymetrix microarray data include: (a) a covariate model, in which the observed signal is some estimated linear function of perfect match (PM) and mismatch (MM) signals; (b) a difference model [PM-MM]; and (c) a PM-only model, in which MM data is not utilized.
Methods: By decomposing the correlations among the variables in the statistical model and making certain assumptions, we theoretically derive the statistical model that reflects the actual gene expression level under a variety of conditions expected in microarray data.
Results and conclusion: When modeling non-systematic variation, the covariate model provides maximum flexibility and often reflects the actual gene expression levels better than the difference model. However, the PM-only model demonstrates superior power in an overwhelming majority of realistic situations, which provides theoretical support for the current trend to employ PM-only models in microarray data analyzes.