S19.1: Association between classical cardiovascular risk factors and age-related maculopathy. First results of the Muenster Age and Retina Study (MARS).
✍ Scribed by Burkhard Dasch; Andrea Fuhs; Thomas Behrens; Daniel Pauleikhoff; Hans-Werner Hense
- Publisher
- John Wiley and Sons
- Year
- 2004
- Tongue
- English
- Weight
- 75 KB
- Volume
- 46
- Category
- Article
- ISSN
- 0323-3847
No coin nor oath required. For personal study only.
✦ Synopsis
The availability of micro array data provides expression profiles based on thousands of genes. For example one goal is to improve diagnosis of different cancer stages by expression arrays of tissue samples. Often machine learning strategies are used to derive such classifiers. Most proposals can be seen as a two step procedure: i) A candidate set of differentially expressed genes is identified by some sort of gene selection. ii) Afterwards supervised classification is applied to a set of genes. Consequently, these proposals focus on additive effects of the genes and allow not for interactions or smaller subgroup effects. I propose and discuss tree like strategies to identify differentially expressed interactions of genes. The talk is illustrated using several micro array data sets.