Methods for assessing reproducibility of clustering patterns observed in analyses of microarray data

McShane, Lisa M.; Radmacher, Michael D.; Freidlin, Boris; Yu, Rong; Li, Ming-Chung; Simon, Richard

doi:10.1093/bioinformatics/18.11.1462

Cited by 194 publications

(159 citation statements)

References 19 publications

Supporting

Mentioning

156

Contrasting

Order By: Relevance

“…On a randomly selected subset of the tumors, a second independent reading was obtained and the agreement between the two readings was assessed using Samples and expression of progenitor cell markers were clustered using hierarchical clustering using Euclidian distance and complete linkage and presented using a heatmap. 17,18 To assess the clustering stability, an R index was calculated by perturbing the data through the introduction of random noise and then re-clustering the perturbed data; this process is repeated multiple times, and the results are compared with the original cluster of unperturbed data. The R index is the proportion of times that a patient pair clusters the same way in the perturbed data sets as in the original data set.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Progenitor stem cell marker expression by pulmonary carcinomas

et al. 2010

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

“…In this study, the perturbed data sets were calculated by introducing a small probability (5%) of flipping the binary marker values rather than by the introduction of Gaussian noise. 18 …”

Section: Discussionmentioning

confidence: 99%

Progenitor stem cell marker expression by pulmonary carcinomas

et al. 2010

View full text Add to dashboard Cite

“…Nevertheless, the stability and reliability of the obtained clusters is crucial to assess the confidence and the significance of a bio-medical discovery [33,34].…”

Section: Introductionmentioning

confidence: 99%

“…Some recent approaches to estimate the reliability of the discovered clusters are based on the concept of the stability with respect to perturbations [33][34][35]. In the context of gene expression data, that are usually characterized by relatively high level of noise [36], stability can be considered an important property: how much the characteristics and composition of the discovered clusters hold when perturbation such as added noise, resampling or random projections are introduced?…”

Section: Introductionmentioning

confidence: 99%

Randomized maps for assessing the reliability of patients clusters in DNA microarray data analyses

Bertoni

Valentini

2006

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

Objective: Clustering algorithms may be applied to the analysis of DNA microarray data to identify novel subgroups that may lead to new taxonomies of diseases defined at bio-molecular level. A major problem related to the identification of biologically meaningful clusters is the assessment of their reliability, since clustering algorithms may find clusters even if no structure is present. Methodology:Recently, methods based on random "perturbations" of the data, such as bootstrapping, noise injections techniques and random subspace methods have been applied to the problem of cluster validity estimation. In this framework, we propose stability measures that exploits the high dimensionality of DNA microarray data and the redundancy of information stored in microarray chips. To this end we randomly project the original gene expression data into lower dimensional subspaces, approximately preserving the distance between the examples according to the Johnson-Lindenstrauss (JL) theory. The stability of the clusters discovered in the original high dimensional space is estimated by comparing them with the clusters discovered in randomly projected lower dimensional subspaces. The proposed cluster-stability measures may be applied to validate and to quantitatively assess the reliability of the clusters obtained by a large class of clustering algorithms. Results and conclusion:We tested the effectiveness of our approach with high dimensional synthetic data, whose distribution is a priori known, showing that the stability measures based on randomized maps correctly predict the number of clusters and the reliability of each individual cluster. Then we showed how to apply the proposed measures to the analysis of DNA microarray data, whose underlying distribution is unknown. We evaluated the validity of clusters discovered by hierarchical clustering algorithms in diffuse large B-cell lymphoma (DLBCL) and malignant melanoma patients, showing that the proposed reliability measures can support bio-medical researchers in the identification of stable clusters of patients and in the discovery of new subtypes of diseases characterized at bio-molecular level.

show abstract

“…The molecular classification in this study did not include the current clinical parameters like tumour grade, steroid receptor status, and HER-2/ neu. In essence, there may be more clusters and molecular subtypes of breast cancer that may be apparent if larger sample sets are available (McShane et al, 2002). Such formal statistical testing has not yet been carried out on the current molecular classification.…”

Section: Expression Microarray Analyses For the Identification Of Promentioning

confidence: 99%

Genomic approaches in the management and treatment of breast cancer

2005

View full text Add to dashboard Cite

Breast cancer is the most common malignancy afflicting women from Western cultures. It has been estimated that approximately 211 000 women will be diagnosed with breast cancer in 2003 in the United States alone, and each year over 40 000 women will die of this disease. Developments in breast cancer molecular and cellular biology research have brought us closer to understanding the genetic basis of this disease. Unfortunately, this information has not yet been incorporated into the routine diagnosis and treatment of breast cancer in the clinic. Recent advancements in microarray technology hold the promise of further increasing our understanding of the complexity and heterogeneity of this disease, and providing new avenues for the prognostication and prediction of breast cancer outcomes. The most recent application of microarray genomic technologies to studying breast cancer will be the focus of this review. Mortality from breast cancer results from the ability of some tumours to metastasise to distant sites. Selecting patients with micrometastases at diagnosis is crucial for clinicians in deciding who should and who should not receive toxic and expensive adjuvant chemotherapy to eradicate these metastatic cells. Axillary nodal status, the best marker now available, is still an imperfect indicator, since about 25% of node-negative patients do harbour micrometastases and are destined to recur, while up to 25% of node-positive patients will not recur even without adjuvant treatment after many years of follow-up. In spite of extensive studies, expression of most individual genes has not proven powerful enough for routine clinical use to predict accurately distant metastases over the lifetime of an individual patient. However, recent developments in applying microarray technologies to breast tumour samples suggest that these new techniques may provide for the transition of molecular biological discoveries to clinical application, and will generate clinically useful genomic profiles that more accurately predict long-term outcome of individual breast cancer patients. NATURAL HISTORY OF BREAST CANCERBreast cancer is characterised by a very heterogeneous clinical course. A major goal of recent studies is to determine whether RNA microarray expression profiling, or DNA array gene amplification/gene loss patterns, can accurately predict an individual's long-term potential for recurrence from breast cancer, so that appropriate treatment decisions can be made. It is well established that some aspects of breast cancer heterogeneity is related to the different risk factors for diagnosis of this disease, such as race, diet, age, environmental factors, and cumulative exposure to the sex hormone oestrogen. The diversity in clinical course of breast cancer is undoubtedly related to differences in tumour growth rates, tumour invasiveness, metastatic potential, and other complex cellular growth signalling and survival pathways. It has long been held that knowledge of these various biological factors would help individualise patient t...

show abstract

Methods for assessing reproducibility of clustering patterns observed in analyses of microarray data

Cited by 194 publications

References 19 publications

Progenitor stem cell marker expression by pulmonary carcinomas

Progenitor stem cell marker expression by pulmonary carcinomas

Randomized maps for assessing the reliability of patients clusters in DNA microarray data analyses

Genomic approaches in the management and treatment of breast cancer

Contact Info

Product

Resources

About