{"title":"Comparison of mixed modeling regression methods for the assessment of longitudinal CyTOF® data","authors":"Tyson H. Holmes, Caroline Duault","doi":"10.1016/j.jim.2025.113880","DOIUrl":null,"url":null,"abstract":"<div><div>A simulation study was conducted to assess Type I error and statistical power of linear mixed models, generalized linear mixed models, and linear quantile mixed models for the analysis of longitudinal cytometry by time-of-flight data. Findings indicate that, while generalized linear mixed models have superior statistical power, they also suffer from inflated Type I error rates. Linear mixed models can have substantially lower statistical power than the other methods at larger effect sizes. While linear quantile mixed models have slightly lower statistical power at smaller effect sizes, they have intermediate statistical power at larger effect sizes and good Type I error control. Taken altogether, these results generally recommend the use of linear quantile mixed models for longitudinal datasets such as for cytometry by time-of-flight data, although linear mixed models may be slightly more useful at small effect sizes and sample sizes.</div></div>","PeriodicalId":16000,"journal":{"name":"Journal of immunological methods","volume":"541 ","pages":"Article 113880"},"PeriodicalIF":1.6000,"publicationDate":"2025-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of immunological methods","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0022175925000808","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"BIOCHEMICAL RESEARCH METHODS","Score":null,"Total":0}
引用次数: 0
Abstract
A simulation study was conducted to assess Type I error and statistical power of linear mixed models, generalized linear mixed models, and linear quantile mixed models for the analysis of longitudinal cytometry by time-of-flight data. Findings indicate that, while generalized linear mixed models have superior statistical power, they also suffer from inflated Type I error rates. Linear mixed models can have substantially lower statistical power than the other methods at larger effect sizes. While linear quantile mixed models have slightly lower statistical power at smaller effect sizes, they have intermediate statistical power at larger effect sizes and good Type I error control. Taken altogether, these results generally recommend the use of linear quantile mixed models for longitudinal datasets such as for cytometry by time-of-flight data, although linear mixed models may be slightly more useful at small effect sizes and sample sizes.
期刊介绍:
The Journal of Immunological Methods is devoted to covering techniques for: (1) Quantitating and detecting antibodies and/or antigens. (2) Purifying immunoglobulins, lymphokines and other molecules of the immune system. (3) Isolating antigens and other substances important in immunological processes. (4) Labelling antigens and antibodies. (5) Localizing antigens and/or antibodies in tissues and cells. (6) Detecting, and fractionating immunocompetent cells. (7) Assaying for cellular immunity. (8) Documenting cell-cell interactions. (9) Initiating immunity and unresponsiveness. (10) Transplanting tissues. (11) Studying items closely related to immunity such as complement, reticuloendothelial system and others. (12) Molecular techniques for studying immune cells and their receptors. (13) Imaging of the immune system. (14) Methods for production or their fragments in eukaryotic and prokaryotic cells.
In addition the journal will publish articles on novel methods for analysing the organization, structure and expression of genes for immunologically important molecules such as immunoglobulins, T cell receptors and accessory molecules involved in antigen recognition, processing and presentation. Submitted full length manuscripts should describe new methods of broad applicability to immunology and not simply the application of an established method to a particular substance - although papers describing such applications may be considered for publication as a short Technical Note. Review articles will also be published by the Journal of Immunological Methods. In general these manuscripts are by solicitation however anyone interested in submitting a review can contact the Reviews Editor and provide an outline of the proposed review.