Querying Co-regulated Genes on Diverse Gene Expression Datasets Via Biclustering.
Methods Mol Biol
; 1375: 55-74, 2016.
Article
en En
| MEDLINE
| ID: mdl-26626937
Rapid development and increasing popularity of gene expression microarrays have resulted in a number of studies on the discovery of co-regulated genes. One important way of discovering such co-regulations is the query-based search since gene co-expressions may indicate a shared role in a biological process. Although there exist promising query-driven search methods adapting clustering, they fail to capture many genes that function in the same biological pathway because microarray datasets are fraught with spurious samples or samples of diverse origin, or the pathways might be regulated under only a subset of samples. On the other hand, a class of clustering algorithms known as biclustering algorithms which simultaneously cluster both the items and their features are useful while analyzing gene expression data, or any data in which items are related in only a subset of their samples. This means that genes need not be related in all samples to be clustered together. Because many genes only interact under specific circumstances, biclustering may recover the relationships that traditional clustering algorithms can easily miss. In this chapter, we briefly summarize the literature using biclustering for querying co-regulated genes. Then we present a novel biclustering approach and evaluate its performance by a thorough experimental analysis.
Palabras clave
Texto completo:
1
Colección:
01-internacional
Base de datos:
MEDLINE
Asunto principal:
Análisis por Conglomerados
/
Biología Computacional
/
Perfilación de la Expresión Génica
Límite:
Humans
Idioma:
En
Revista:
Methods Mol Biol
Asunto de la revista:
BIOLOGIA MOLECULAR
Año:
2016
Tipo del documento:
Article
País de afiliación:
Estados Unidos
Pais de publicación:
Estados Unidos