Joint position with the Mathematical Institute of Leiden University.
My work focuses on prediction from omics data. Often extra information on the variables in omics data is available. An example from genomics is the location of a gene on the genome. While this information can rarely be directly incorporated in the statistical analysis it may still be of predictive value. I work on penalized regression methods with multiple penalty parameters that model the external information. Rather than fixing the weights of these external data, we estimate their contribution.
gren: a group-regularized logistic elastic net package to include external information in high-dimensional prediction.
NIG: Jointly estimates a set of related univariate Bayesian regression problems.
Münch, M.M., van de Wiel, M.A., Richardson, S., & Leday, G.G.R. (2020). Drug sensitivity prediction with normal inverse Gaussian shrinkage informed by external data. Biometrical Journal, to appear.
Hu, W., Yang, S., Shimada, Y., Münch, M., Marín-Juez, R., Meijer, A. H., & Spaink, H. P. (2019). Infection and RNA-seq analysis of a zebrafish tlr2 mutant shows a broad function of this toll-like receptor in transcriptional and metabolic control and defense to Mycobacterium marinum infection. BMC Genomics, 20, 878. article
Van de Wiel, M.A., te Beest, D.E., & Münch, M.M. (2018). Learning from a lot: Empirical Bayes in high-dimensional prediction settings. Scandinavian Journal of Statistics, 1-24, doi:10.1111/sjos.12335. article