mixOmics is an R package dedicated to the statistical analyses of large biological data sets such as ‘omics’ data or highly dimensional data sets, where the number of measured entities (measurements of gene, metabolite expression, protein abundance) is much larger than the number of samples (the number of patients, for example). mixOmics provides a strong focus on data visualisation in order to better interpret the results.
Several methodologies are implemented in mixOmics, particularly to integrate two data sets. Canonical Correlation Analysis, Partial Least Squares have been further improved to deal with the high dimension of the data (regularised CCA) or to select relevant variables that are correlated within and across the two data sets (sparse Partial Least Squares).
The flexibility of mixOmics also proposes a large range of tools to explore highly dimensional data sets, such as Principal Component Analysis, Linear Discriminant Analysis and sparse Linear Discriminant Analysis., and recently, Principal Components with Independent Loadings (IPCA). All methodologies implemented in mixOmics produce the same kind of graphical output to facilitate the interpretation of the results.
mixOmics is in the process of being improved and more methodologies are currently being developed in collaboration with the Université de Toulouse. An example is identifying correlated profiles in longitudinal studies or to integrate more than two data sets.
To subscribe to our newsletter, simply send an e-mail with no subject or body to this address.
This project acknowledges the support of the Australian Government’s Cooperative Research Centres Program.