Adaptive quality-based clustering of gene expression profiles

Smet, Frank De; Mathys, Janick; Marchal, Kathleen; Thijs, Gert; Moor, Bart De; Moreau, Yves

doi:10.1093/bioinformatics/18.5.735

Cited by 170 publications

(118 citation statements)

References 27 publications

Supporting

Mentioning

116

Contrasting

Unclassified

Order By: Relevance

“…Spot intensities were normalized using a Loess-fit (13) for removing nonlinear dye related variation followed by a global analysis of variance normalization (14). The obtained expression data were clustered with the AQBC algorithm (15).…”

Section: Methodsmentioning

confidence: 99%

Characterization of the Condensin Component Cnap1 and Protein Kinase Melk as Novel E2F Target Genes Down-regulated by 1,25-Dihydroxyvitamin D3

Verlinden

Eelen

Beullens

et al. 2005

Journal of Biological Chemistry

View full text Add to dashboard Cite

Active complexes between cyclin D and cyclin-dependent kinases 4/6 regulate the transition through the G 1 /S restriction point by phosphorylation of the retinoblastoma protein (pRb) 3 and other members of the pocket protein family, p107 and p130. The phosphorylation status of these pocket proteins determines their association with members of the E2F family of transcriptional regulators, which play a pivotal role in mediating gene expression during cell proliferation. These E2F proteins can be allocated to four subclasses. Upon release by their pocket protein pRb, E2Fs 1-3 function as transcriptional activators in late G 1 and in S phase. E2F4 and E2F5 act as transcriptional repressors in quiescent and early G 1 cells by associating with p107 or p130 (1, 2). In quiescent cells repression of the promoter activity of E2F target genes is associated with the recruitment of E2F4 and p130 and low levels of histone acetylation. By late G 1 , these proteins are largely replaced by activator E2Fs in concert with histone acetylation and gene activation. It is, therefore, likely that two pathways, one controlled by pRb and the other by p130/p107, regulate distinct downstream events required for G 1 progression and G 1 /S transition (2, 3). Recently, the transcriptional repressor E2F6 was proposed to make up the third subclass of E2F proteins (4, 5), whereas E2F7 and E2F8 form the last subclass and are thought to regulate a subset of E2F target genes during the cell cycle (6, 7).1,25-Dihydroxyvitamin D 3 (1,25(OH) 2 D 3 ), the active metabolite of vitamin D 3 , acts on bone and mineral homeostasis and also inhibits proliferation and induces differentiation of various normal and malignant cells (8). However, the exact molecular mechanism behind this growth-inhibitory effect is unknown. 1,25(OH) 2 D 3 has a cell cycle-specific effect leading to an accumulation of cells in the G 1 phase of the cell cycle (9). It has been shown previously that 1,25(OH) 2 D 3 reduces the activity of the cyclin D1-cyclin-dependent kinase 4/6 complex, which may contribute to its antiproliferative effect (10).In the present study a cDNA microarray was performed to examine the expression profile of 21,492 genes in MC3T3-E1 cells treated with 1,25(OH) 2 D 3 for different times up to 36 h. Statistical analysis revealed a cluster of down-regulated genes involved in cell cycle regulation and in DNA replication but also in checkpoint control, DNA repair, chromosome transactions, and mitosis. Approximately 30% of the genes in this cluster are known E2F targets, and in silico promoter analysis demonstrated an additional 20% of the genes to contain E2F binding sites in their promoter. Four of these genes were selected for further analysis, namely Cnap1, Melk, retroviral integration site 2 (Ris2), and enhancer of Zeste homolog 2 (Ezh2). Expression of these genes was growth-regulated as were the promoter activities of Cnap1 and Melk. Mutational analysis revealed that the identified E2F binding sites were required for transactivation by E2F family members. R...

show abstract

Section: Methodsmentioning

confidence: 99%

Characterization of the Condensin Component Cnap1 and Protein Kinase Melk as Novel E2F Target Genes Down-regulated by 1,25-Dihydroxyvitamin D3

Verlinden

Eelen

Beullens

et al. 2005

Journal of Biological Chemistry

View full text Add to dashboard Cite

show abstract

“…Most existing algorithms require this parameter to be provided, which is a major problem in a setting where the structure of the data is completely unknown. Whilst there have been recent attempts to determine the number of clusters automatically (including the Gap statistic , Resampling (Dudoit and Fridlyand, 2002) and others (De Smet et al, 2002)), no entirely reliable method exists to date.…”

Section: Clustering Methodsmentioning

confidence: 99%

A metabolome pipeline: from concept to data to knowledge

et al. 2005

View full text Add to dashboard Cite

Metabolomics, like other omics methods, produces huge datasets of biological variables, often accompanied by the necessary metadata. However, regardless of the form in which these are produced they are merely the ground substance for assisting us in answering biological questions. In this short tutorial review and position paper we seek to set out some of the elements of ''best practice'' in the optimal acquisition of such data, and in the means by which they may be turned into reliable knowledge. Many of these steps involve the solution of what amount to combinatorial optimization problems, and methods developed for these, especially those based on evolutionary computing, are proving valuable. This is done in terms of a ''pipeline'' that goes from the design of good experiments, through instrumental optimization, data storage and manipulation, the chemometric data processing methods in common use, and the necessary means of validation and cross-validation for giving conclusions that are credible and likely to be robust when applied in comparable circumstances to samples not used in their generation.

show abstract

“…These algorithms include, among others, model-based algorithms [133,134], the self-organizing tree algorithm [135], quality-based algorithms [136]-which produce clusters with a quality guarantee that ensures that all members of a cluster are co-expressed-and biclustering algorithms [137]-they cluster both the genes and the experiments at the same time.…”

Section: Clustering In Bioinformaticsmentioning

confidence: 99%