Gene clustering at intensice science: comparison using biological stability index
Abstract This research evaluates the performance of the best-known clustering algorithms using the biological stability index (BSI). A comparison was made between the clustering algorithms, to determine which is the optimum according to the score obtained in each algorithm, the group of genetics in Intensive Science, which uses extensive databases to cover almost all the results that could probably really. This method is applied to a gene expression database (Microarray). The analysis was performed on the "mouse" database included in the clValid package in the R software, for the study of mouse mesenchymal cells (neural crest and derived mesoderm), graphic methods, such as dendograms, are used for a first approach. For the selection of the optimal algorithm, the biological stability index was calculated for each clustering algorithm, the best being the one closest to the unit. Consequently, the most stable algorithm for this database is "Diana". To reach this result, the number of clusters with the response obtained in each case was visualized graphically; the optimal algorithm was taken as the one that most closely matches the reality of the problem, taking into account its score in the indexes and also with the help of a phylogenetic graph for a final approach.
| Main Authors: | , , |
|---|---|
| Format: | Digital revista |
| Language: | English |
| Published: |
Escuela Superior Politécnica de Chimborazo
2020
|
| Online Access: | http://scielo.senescyt.gob.ec/scielo.php?script=sci_arttext&pid=S2477-91052020000200012 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|