Algorithms for clustering data
Algorithms for clustering data
Self-organizing maps
The gene expression matrix: towards the extraction of genetic network architectures
Proceedings of the second world congress on Nonlinear analysts: part 3
IPCAT '97 Proceedings of the second international workshop on Information processing in cell and tissues
Tissue classification with gene expression profiles
RECOMB '00 Proceedings of the fourth annual international conference on Computational molecular biology
Rough Sets in Oligonucleotide Microarray Data Analysis
RSEISP '07 Proceedings of the international conference on Rough Sets and Intelligent Systems Paradigms
Information Sciences: an International Journal
Hi-index | 0.00 |
The goal of many gene expression experiments is to discover genes that are functionally related by clustering expression levels sampled over some time interval, with the hope that co-regulated genes also are functionally related. However, it is not necessarily always true that co-regulated genes are functionally related, or vice versa, and therefore this paper investigates the value of including gene annotation in the clustering process. Results suggest that clusters formed by a clustering of a combination of expression data and annotation in the form of enzyme classification can give results that have higher correlation with known biological data (functional and metabolic pathway) not included in the clustering process. The results show that the same is true even in a situation with only 10% of the dataset annotated, which is an estimate of the amount of enzymatic annotation available today and a sign that the inclusion of added data helps in the clustering of genes without any explicit annotation.