Do you know a method that could be used to decide the optimum results (number of groups) when performing clustering analysis?
A simple way is to select a distance where the branches are longer, but depending of the tree this can be difficult. In that case there are some computational methods such as the Dynamic Cut Method that process the topology of the tree (http://bioinformatics.oxfordjournals.org/content/24/5/719.full). There is a R package to use this methodology.
Estimada Mamen, le aconsejo usar el paquete JMP
Este paquete da el punto de corte
Saludos
Emmanuel
Hi
This paper may be useful for your purpose: Suzuki, R, Shimodaira, H. (2006), Pvclust: an R package for assessing the uncertainty in hierarchical clustering, Bioinformatics, 22 (12), 1540-1542
Best,
Gabriele
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology