Research Seminar
Sparse simultaneous component analysisKatrijn Van DeunKU Leuven | |
| Abstract: | Integrating several data sets that are obtained on the same set of samples but that are derived from different sources is one of the major challenges in systems biology. Often the aim of such a data integration is to find a few important biological processes that underlie the data. Simultaneous component methods have been shown to be most useful in this regard. Interpretation of the simultaneous components is based on the weights of the variables on the components. In systems biology, however, the number of variables is often large (e.g., in the case of omics data). To facilitate the interpretation, we propose a sparse simultaneous component method that shrinks a tunable number of variable weights to zero. In this presentation, this method and an associated algorithm will be proposed. We will illustrate with an application on metabolome data obtained on 28 samples of Escherichia coli screened with two different mass spectrometry methods. |
| Date: | Tue Oct 26, 12:15 pm - 1:15 pm |
| Place: | room 00.98 (Department of Psychology, Tiensestraat 102, 3000 Leuven) |