The most powerful multivariate normality test for plant genomics and dynamics data sets

Data analysis methods like analysis of variance and regression in plant sciences depend on the assumption that the biological data are normal. Using a normality test is the best way to check whether the distribution is normal or not. Plant genomic and dynamic studies generate data with leptokurtic distribution and the most appropriate normality test is the Shapiro–Francia one. However multivariate extensions of this test have not been designed yet and plant data matrix cannot be performed efficiently or without bias. Thus, our analysis focused on the development of an easy-using algorithm to extend the application of the Shapiro–Francia test to multivariate data matrix in plant studies.