Purpose: Cytokine data sets are increasing both in the number of different cytokines measured and the number of samples assayed. Further, typically data from different groups may be contrasted, eg, normal vs complication subjects. Many univariate and multivariate statistical techniques exist to study such cytokine datasets, but the ability to implement these techniques may be lacking for some practitioners, or may not be available quickly and conveniently. Here, we introduce CytokineExplore, an online tool for multi-cytokine and multi-group data analysis of user-provided Microsoft Excel data files.
Materials and methods: In order to illustrate the tool features, we use data from intrauterine growth retardation (IUGR), a pregnancy complication, and normal healthy subjects as a control. The dataset contains levels for 10 cytokines, namely: IL-4, IL-6, IL-8, IL-10, IL-12, IL-13, IL-18, IL-23, interferon-gamma (IFN-γ) and tumour necrosis-alpha (TNF-α), obtained from 34 women with IUGR (further divided into 17 symmetric and 17 asymmetric cases) and 24 gestationally age-matched normal controls.
Results: The online tool automatically generates box-plots, histograms, PCA and PLSDA plots, t-tests and Mann-Whitney statistical tests, cytokine importance values for separating two groups, heatmaps for comparing multiple groups, and other functionalities. Figures generated can be directly downloaded for use in presentations or journal articles.
Conclusion: The tool facilitates quick and easy numerical exploration and multivariate analysis of cytokine datasets, to aid basic understanding and hypothesis generation.
Keywords: cytokines; data analysis; diagnostics; intrauterine growth retardation; multivariate statistics; pregnancy complications.
© 2020 Kassem et al.