MVAPP – Multivariate analysis application for streamlined data analysis and curation
journal contributionposted on 20.05.2018, 06:44 by Magdalena JulkowskaMagdalena Julkowska, Stephanie Saade, Gaurav Agarwal, Ge Gao, Yveline Pailles, Mitchell Morton, Mariam Awlia, Mark Tester
Modern phenotyping enables the measurement of many phenotypic traits simultaneously, yielding vast amounts of quantitative data that is hard to manage and analyze. This type of data, when adequately examined, could reveal genotype-to-phenotype relationships and meaningful relationships between individual measured traits. Efficient data mining is currently challenging for experimental biologists, as many researchers are limited by their ability to curate, integrate and explore these complex outputs. Additionally, data transparency, accessibility and reproducibility have become important considerations for scientific publication. Thus, the need for a streamlined pipeline for curating phenotypic data is now more pressing than in the past. To address that need, we developed an open-source online platform for multivariate analysis, MVApp, which allows interactive data curation, in-depth data analysis and customized visualization. MVApp was developed in R using the Shiny framework, combining several functional R modules into a comprehensive toolkit that can be used without any prior knowledge of R, programming or advanced statistics. MVApp aims to enhance data transparency, standardize and facilitate phenotypic data curation and increase statistical literacy among the scientific community. Given that contributions from users in this open-source environment is encouraged, MVApp can be continuously updated and expanded to facilitate the analysis of high-throughput phenotyping outputs.
This work was funded by King Abdullah University of Science and Technology (KAUST), Kingdom of Saudi Arabia
Multivariate AnalysisMVAppData analysisoutlier removalnormal distributionequal variancecurve fittingdescriptive modelingpolynomial curvesANOVAtwo-way ANOVAt-testcorrelation analysesPrincipal component analysisMultidimensional scalingquantile regressionk-means clusteringhierarchical clusteringdimensionality reductionbroad-sense heritabilitydata analysis transparency